Evolution, Genetics, and Man

Evolution, Genetics,

and Man

THEODOSIUS 'DOBZHANSKY

PROFESSOR OF ZOOLOGY AND MEMBER OF THE

INSTITUTE FOR THE STUDY OF HUMAN VARIATION

COLUMBIA UNIVERSITY

Evolution, Genetics

^i

and Man

JOHN WILEY & SONS, INC. - NEW YORK

CHAPMAN & HALL, LIMITED - LONDON

FOURTH PRINTING, OCTOBER, 1959

Copyright © 1955

by

John Wiley & Sons, Inc.

All Rights Reserved

This book or any part thereof must not

be reproduced in any form without the

written permission of the publisher.

Library of Congress Catalog Card Number: 55-10868

Printed in the United States of America

Preface

For almost a century the influence of the theory of evolution has

been felt far beyond the limits of biology. In fact, this influence has

been growing apace, and in our day the idea of evolution has become

an integral part of the intellectual equipment of Western Civilization.

In biology this idea is pivotal. To a beginning student no less than

to a teacher and to a specialist, the idea of evolution makes sense

of what would otherwise be wearisome descriptions of arid facts to

be memorized, only to be forgotten as soon as the course is over. In

the light of evolution the same facts and descriptions of creatures

which we have seldom or never seen become fascinating. Learning

them turns out to be an intellectual adventure.

Biological evolution is a part of the evolution of the cosmos. The

rise and the development of mankind are a part of the story of bio-

logical evolution. Man cannot reach a valid understanding of his

own nature without a knowledge of his own biological background.

It may, then, be that the study of evolutionary biology is the most

important practical endeavor open to the human mind. Accordingly,

an effort is being made in this book to show to the student that biology

is not only a craft which is interesting to technicians and devotees but

also a part of the fabric of modern humanistic thought. I am quite

conscious that this goal is too ambitious and that it has not been

fully attained.

I hope that this book may be useful not only as a guide in courses

devoted to the study of biological evolution but also as collateral

reading in courses of general biology, general zoology, general botany,

and anthropology. A sizable proportion of the space in this book is

devoted to presentation of material which is usually given in ele-

mentary courses of genetics. Chapters 2, 3, 4, parts of 5 and 11, and

smaller parts of other chapters deal with genetics. This emphasis is

unavoidable, since modern evolutionism is incomprehensible except

on the basis of familiarity with fundamentals of genetics. Therefore,

vii

viii Preface

although it has not been my intention to turn this book into a brief

textbook of genetics, a student who uses it will acquire an elementary

knowledge of the subject. The book will probably fit the require-

ments of courses given in some institutions of higher learning, courses

entitled "Genetics and Evolution" or "Evolution and Genetics."

An effort has been made to use, wherever possible, examples deal-

ing with man and to point out the bearing of the topics discussed

on human problems. The opinion once held fairly widely, that man

is most unfavorable as material for biological and especially for genetic

studies, is becoming less and less prevalent. Even though we cannot

arrange many genetical experiments with man, there is an abundance

of kinds of information bearing on the genetics and evolution of man

which are not available for any other organism. And after all is said

and done, the species Homo sapiens happens to be more interesting

to most students than any other species, no matter how unserviceable

it may be for some experiments. Being men, we understand many

biological phenomena in man more easily and more clearly than we

do the biology of much "simpler" organisms.

Although this book is meant to be comprehensible to a student with

no more than an elementary previous knowledge of biology (at about

high school level), some more "advanced" material and discussion of a

number of unsettled and controversial problems have been included.

As a result, the book contains more material than can be adequately

covered in an average one-term course; but this superabundance of

material is deliberate. The subdivision of the chapters into short

sections with descriptive subheadings should make deletion of the

unwanted material easy. On the other hand, what can be more chal-

lenging and inspiring to a student of average and above-average in-

telligence than to learn that science is not something all completed

and finished, merely to be memorized from books, but a growing body

of knowledge, in the development of which this same student may

have a hand if he so chooses? Is finding this out not equally, or even

more, valuable to a student than learning more "facts"?

In place of a conventional chapter on the history of evolutionary

theories, the history of various ideas and concepts is discussed in this

book in the same chapters which present the modern status of those

ideas and concepts. This arrangement of material does not in any

sense mean an underestimate of the importance of the history of sci-

ence or of its interest to an intelligent student. But modern evolu-

tionary thought is a result of confluence and integration of the work

of many biological disciplines, which even in a recent past were de-

Preface ix

veloping more independently than they are now. The history of the

evolutionary doctrine as a whole, from a modern standpoint, has never

been written, and for the time being it seems more convenient to

present the historical information piecemeal.

The "Suggestions for Further Reading" given at the end of most

chapters are meant to assist the student who may wish to go beyond

the limits of this book in exploring problems of evolutionary biology.

These "Suggestions" are certainly not meant to serve as bibliographies

in which a reference to an authority for every fact and name men-

tioned in the text may be located.

I am deeply indebted to several colleagues and friends who have

read chapters of this book and suggested corrections and improve-

ments. The greatest thanks go to Professors Ernst Mayr, of Harvard

University, Charles Birch, of the University of Sydney, Australia, and

A. B. da Cunha, of the University of Sao Paulo, Brazil, who have read

the manuscript in its entirety. Chapters 1, 2, and 3 were read also by

Drs. Alfred Mirsky and Stanley Gartler; Chapters 4 and 5 by Dr. M.

Demerec; Chapters 6 and 7 by Dr. Phillip M. Sheppard; Chapters 6, 7,

8, and 10 by Professor H. L. Carson; Chapters 8, 10, and 12 by Pro-

fessor John A. Moore; Chapter 9 by Professor P. C. Mangelsdorf; Chap-

ter 11 by Professor Aubrey Gorbman; and Chapters 13 and 14 by Pro-

fessor L. C. Dunn, Mr. Stephen Dunn, and Mr. M. D. Coe. Quite

obviously, I remain solely responsible for all errors of commission and

omission which doubtless will be found in the book. Mr. Stephen R.

Peck has drawn many of the excellent illustrations which adorn the

text. Several colleagues have contributed other illustrations and photo-

graphs, as acknowledged in the legends to these figures. Finally,

thanks are due Miss Adelaide Richardson, who prepared the typescript.

I'll. DOBZHANSKY

Sao Paulo, Brazil

October, 1955

Contents

1 • Nature and Origin of Life 1

2- The Gene as the Basis of Evolution <—•~" 23

3- Chromosomes as Gene Carriers ^ 44

'4- Heredity, Environment, and Mutation 72

. 5- Elementary Evolutionary Changes or Microevolution 91

6- Natural Selection and Adaptation 109

' 7 • Individuals, Populations, and Races 134

8- Species 165

9- Evolution under Domestication and Evolution by

Polyploidy 191

10- Evolution of the Organic Form and Function 222

11 • Evolution of Sex 253

12- Historical Record of Organic Evolution 284

13- Human Evolution 319-

^14- Chance, Guidance, and Freedom in Evolution 357

Index 381

Nature and Origin

of Life

The purpose of science is twofold. Science strives, in the first place,

to understand man and the universe of which he is a part. In the sec-

ond place, science endeavors to provide man with the means to con-

trol his environment. The quest for understanding is a function of

theoretical, fundamental, or pure science. Knowledge and under-

standing are sources of satisfaction even when they do not yield any

immediate material benefits. Control of the environment is a func-

tion of applied science or technology.

Understanding things, however, is the surest approach to controlling

them; and the distinction between pure and applied science is, there-

fore, not always sharp. This distinction often describes the attitudes

of mind of investigators and students rather than the subject matter

of their investigations and studies. Some discoveries of greatest prac-

tical utility have been made by scientists engaged in exploration of

the laws of nature without regard for their possible utilization. For in-

stance, the germ theory of disease and much of the modern food tech-

nology are outgrowths of the studies of the great French biologist

Pasteur (1822-1895) on the nature of life.

Cosmic Evolution. Discoveries made in various branches of science

during the nineteenth and twentieth centuries have converged to estab-

lish an evolutionary approach to the understanding of nature. The

universe has not always been as it is now. Nature as we observe it

today is the outcome of a historical process of development, evolution.

The human race with its social, intellectual, and artistic achievements,

the world of living creatures, and inanimate nature, all evolved gradu-

ally and by stages from very different antecedents.

The classical atomist view of nature, which dominated physical sci-

1

2 Nature and Origin of Life

ences up to the beginning of the current century, held that the basic

physical and chemical properties of matter were always as they are

today. Matter consists of atoms of several scores of chemical elements,

and the atoms combine with each other according to certain rules

which the chemists describe. All material objects in the universe con-

sist of different combinations of atoms, but the atoms themselves were

believed to be unchangeable and indivisible. The very word "atom"

means indivisible in Greek.

Classical atomism was right as far as it went, but it oversimplified

the actual situation. Physicists have shown that atoms consist of still

smaller units—electrons, neutrons, protons. Atoms of chemical ele-

ments have been transformed in laboratory experiments into other

elements. Moreover, modern cosmology, the study of the universe,

assumes that the atoms which exist today have had a history. One of

the theories is that the atoms were formed from a primordial sub-

stance, called the "ylem," and the inference is that they were formed

in a tremendous explosion, which occurred supposedly more than 5

billion years ago. This stupendous event is, then, the first discernible

date in the history of the universe and the beginning of cosmic evolu-

tion. After the formation of the atoms, cosmic evolution led to the

concentration of the original cloud of atoms into galaxies. This proc-

ess took a relatively "short" time, some 30 million years. The stars

and planets were formed within the galaxies. Our earth came into

existence presumably as one of these planets; it is, therefore, only a

little "younger" than the universe itself.

The work of geologists has shown that the earth underwent many

transformations during its long history. Again and again mountain

ranges rose above the plains, and were leveled back by erosion; por-

tions of land sank beneath the seas, and sea bottoms rose to become

land; the climates of many parts of the world changed from warm

to cold, and vice versa.

Biological Evolution. Our earth is an insignificantly small particle

of the universe, yet we cannot be sure that life exists anywhere ex-

cept on this small particle. The evolution of life, biological evolution,

has, to our knowledge, been enacted on earth alone. How long ago

life appeared on earth, however, is a problem fraught with uncertain-

ties. TJie -mostancient indisputable and abundant fossil remains are

estimated to be some 500 million years old. These ancient (Cambrian,

see Chapter 12) fossils are remains of creatures which inhabited the

seas. Because the principal types, or phyla, of the now-existing ma-

rine organisms are represented among Cambrian fossils we can sup-

Human Evolution and Evolution of Culture 3

pose that much evolution took place before these organisms could

have appeared, and, therefore, that the origin of life took place long

before then.

The extreme scarcity of the fossil record of the beginning stages of

biological evolution is probably due to two causes. First, the most

primitive organisms now living are too small and too delicate to be

preserved as fossils, and this was most likely true in the early stages

of organic evolution. Second, the geological strata older than 500

million years (and many of those younger than that) were altered by

heat and by great pressures in such ways that whatever fossils had

been present in them were destroyed. For all these reasons it is

conjectured that life appeared on earth much earlier than 500 million

years ago. Unaltered pre-Cambrian rocks are rare, and they contain

a few doubtful fossils, which some authorities interpret as seaweeds

and algae but others regard as possibly formed without participation

of life. Perhaps more hopeful are very ancient deposits of carbon

apparently of organic origin. Holmes claimed in 1954 that one such

deposit in Africa is between 2.6 and 2.7 billion years old. If con-

firmed, this claim will mean that organic evolution became super-

imposed on cosmic evolution very long ago (Table 1.1).

The first land-growing plants appeared, as shown by the fossil rec-

ord, at least 400 million years ago. Land animals were added later-

some 300 million years ago. The first known land-inhabiting verte-

brate animals are still more recent—200 to 250 million years. Mam-

mals, the class of animals to which man belongs, were evolved some

125 million years ago and became diversified and widespread at most

75 million years ago. Mankind is a newcomer even among mammals.

.The first traces of man's presence are less than one million years old,

which is less than one per cent of the time span during which mam-

mals are known to have lived.

Human Evolution and Evolution of Culture. With the appearance

of man a third kind of evolution, that of the human spirit, became

superimposed on the background of the biological and cosmic evolu-

tion. Of course the entry of man on the evolutionary stage did not

mean that biological evolution had come to an end, no more than the

origin of life meant the termination of cosmic evolution. The three

kinds of evolution proceeded at the same time.

The Greek Anaximander (611-547 B.C.), the first evolutionist to

leave a trace in the history of human thought, taught that life arose

from mud warmed up by sun rays. Plantj^cajpe first, then animals,

and finally man. But, reasoned Anaximander, man could not have

4 Nature and Origin of Life

TABLE 1.1

SOME APPROXIMATE DATES OF COSMIC, BIOLOGICAL, AND HUMAN EVOLUTION

Years Ago

(Approximate) Events

100 Publication of Darwin's Origin of Species

100-200 Industrial Revolution

300-400 Life of Galileo and birth of modern science

1955 Birth of Christ

3000 Beginning of Greek civilization

3500 Beginning of Chinese civilization

5400 Beginning of civilization in Mesopotamia

5400 Beginning of First Dynasty in Egypt

6200 Introduction of Calendar in Egypt

15-25 thousand Man in America

20-50 thousand Cro-Magnon man in Europe

20-75 thousand Old Stone Age in Europe

500-800 thousand First man-made tools

75 million Beginning of the Age of Mammals

125 million The first mammals appear

500 million Beginning of the fossil record

2.5 billion Appearance of life on earth

5 billion Appearance of atoms of the chemical ele-

ments, followed by the formation of the

galaxies, stars, and planets

arisen directly from mud, since as a child he is unable to feed or to

take care of himself. Hence he must have arisen from another ani-

mal. This, then, is the first statement of the view that man is biologi-

cally unique (see Chapter 13). At present we are confident that man

is a product of biological evolution; his evolution was brought about

by the same fundamental causes as the evolution of all other organ-

isms. But in man the biological evolution has transcended itself.

Man is able to use language symbols, to arrive at decisions by a

process of abstraction and reasoning, and to distinguish between good

and evil. Children inherit their biological heredity from their par-

ents through the sex cells, but they inherit their culture by learning

from other human beings, not necessarily related to them by descent.

Biological heredity is set at fertilization, and it remains more or

less constant thereafter (Chapter 6). Cultural heredity is acquired

throughout life, but principally during childhood and youth (Chap-

ters 13 and 14).

To develop culture, the human species had to evolve a human bio-

logical organization. No animal, not even the anthropoid apes, can

Human Evolution and Evolution of Culture 5

acquire the rudiments of human culture. The important thing, how-

ever, is that the biological organization which enabled man to acquire

and develop culture has conferred on him, for that reason, an im-

mense biological advantage (Chapter 13). Man adapts himself to

his environment chiefly by using his technical skills, his knowledge

of things, his science, art, religion, in brief his culture. Now, as indi-

cated above, the process of transmission of culture is vastly more

efficient than biological heredity, which comes only from our parents

and other direct ancestors, and can be transmitted only to our off-

spring. By contrast, learning, art, belief, or wisdom can be trans-

mitted by precept, by speech, or by writing, to any number of human

beings, regardless of their being related by descent or not. Every

one of us has "inherited" the wisdom of people whom we never met

in the flesh. In many instances these people died centuries before

we were born (see Chapter 13).

The rise of man from the animal level to truly human estate was

slow at first. A few bone fragments of a creature which combined

some human-like and some ape-like features were discovered in South

Africa, together with charcoal remains. This discovery led the dis-

coverer, Dart, to surmise that the creature was a user of fire, and to

name it Australopithecus prometheus (pithecus, ape; Prometheus, the

discoverer of fire). The dating of this fossil is, most unfortunately,

quite uncertain; the creature may have lived half a million to more

than a million years ago. Dart's interpretation of his find is regarded

as doubtful.

There is, however, no doubt that at least 20,000 years ago there

appeared in Europe a race of people who, judging by their bones,

might have looked pretty much like ourselves. The drawings of ani-

mals which they made on the walls of the caves which they inhabited

on the territory of the present France show that they possessed an

exquisite artistic feeling (Figure 1.1).

The first light of recorded history dawned in the valley of the Nile,

in Egypt, some sixty-two centuries ago. Within a few centuries a

cultural awakening took place in several countries—in Egypt, in

Mesopotamia (Iraq), in India, and somewhat later in China. Despite

the numerous, and often grievous, setbacks, the development of human

cultures has proceeded since then with, seemingly, accelerating tempo.

To a philosopher the cosmic, biological, and cultural evolutions are

integral parts of the grand drama of Creation. A scientist, though he

recognizes the unity of the evolutionary process, must perforce con-

fine his studies within narrower bounds because the methods of in-

Characteristics of Living Matter 1

A chemical composition including proteins and nucleic acids.

A definite organization.

Maintenance and growth through assimilation.

Reproduction and heredity.

Irritability and adaptation.

The living bodies consist very largely of oxygen, carbon, hydrogen,

and nitrogen, that is, chemical elements which are quite common also

in the inorganic nature. To those are added a number of other com-

mon elements, as shown in the following table, which indicates the

percentages of the various elements in the human body:

Element Per Cent Element Per Cent

Oxygen 65 Potassium 0.35

Carbon 18 Sulphur 0.25

Hydrogen 10 Sodium 0.15

Nitrogen 3 Chlorine 0.15

Calcium 2 Magnesium 0.05

Phosphorus 1 All others 0.05

Some organisms contain rather higher proportions of chemical ele-

ments which are present only as traces in the human frame. Thus

the blood of many mollusks has a greenish color owing to the pres-

ence of copper, and the ascidians (a group of marine animals) con-

tain some vanadium. But there is certainly no chemical element that

would be confined to living protoplasm. Carbon comes nearest to

deserving to be called "the stuff of life," because it is capable of form-

ing innumerable complex compounds with other elements which occur

in living matter. These "organic" compounds are far more char-

acteristic of life than the elements which enter into the composition

of living bodies. Proteins and nucleic acids are the two classes of such

compounds which are important, because they are universally present

in all life. But we cannot be entirely sure that some very different

classes of chemical compounds could not produce life.

Proteins are large, and often huge, molecules, with molecular weights

usually in tens or in hundreds of thousands, and often in millions.

The proteins consist, in turn, of chemically linked building blocks

known as amino acids. About twenty different amino acids are com-

monly found, all of them characterized by the presence of "amino

groups" NH2 and a carboxyl group, COOH, in their molecular makeup.

Their molecular weights are in hundreds. An important property of

amino acids is their ability to combine in a great variety of complex

8 Nature and Origin of Life

patterns such as are found in proteins. The specificity and constancy

of the organization of living beings (see below) is presumed to rest

on the specificity of the proteins. Enzymes, chemical substances which

mediate many important chemical reactions taking place in living

bodies, are among the proteins. The action of enzymes is as a rule

highly specific since each enzyme mediates one and only one kind of

reaction.

Nucleic acids are compounds of so-called nucleotides. A nucleotide,

in turn, consists of a nitrogen-containing purine or pyrimidine base,

a sugar, and a phosphoric acid. Depending on the kind of sugar in-

volved, two types of nucleic acids can be distinguished, called, re-

spectively, ribonucleic acids (RNA, for short) and desoxyribonucleic

acids (or DNA). The DNA is invariably present in the chromosomes

of the cell nucleus, whereas RNA is a characteristic constituent of the

cell cytoplasm. The nucleotides are, in living cells, associated with

each other to form compounds of, often, very high molecular weight.

Furthermore, nucleic acids link up with proteins to form nucleopro-

teins. It appears that nucleoproteins are present in all existing living

bodies, down to the simplest viruses.

Organization and Individuality. There is no question but that com-

prehension of the chemical processes which occur in living organisms

is essential for an understanding of life, and that our knowledge of

these processes is as yet insufficient. But we should not think of an

organism as though it were simply a mixture of chemical substances.

Life occurs always in discrete units, in individuals, which possess a

fairly constant and usually highly complex structure or organization.

From men, elephants, and whales to insects, polyps, and lowly amoe-

bae, and from pine and oak trees to grasses, algae, and down to the

simplest organisms, we can always discern the characteristic external

structures (morphology), internal gross and microscopic anatomy, and

the physiological properties of individuals of every species of or-

ganisms.

Inanimate objects do not usually possess definable individuality.

Mountain ranges, rocks, rivers, or seas are not discrete individuals,

since they do not have a cohesive structure that would be character-

istic also of other individuals of a species. To be sure, crystals seem

to foreshadow on the inorganic level the discreteness and the constant

organization of living individuals. The shape of crystals is fixed by

their chemical structure; the atoms inside a crystal are arranged in a

definite pattern. Under proper conditions a crystal can grow and

even restore missing parts, which makes us think of regeneration of

12 Nature and Origin of Life

of different men but also of other species of animals, for examples of

dogs and cats kept as domestic pets. It is the heredity which causes

the processes of assimilation to result in self-reproduction.

The production of a likeness of the assimilating body and of its

ancestors is most strikingly apparent when the organism grows or

gives rise to a progeny. An oak tree brings forth acorns, each able to

grow into an oak. An oyster gives some hundred million eggs in a

single spawning, each potentially a new oyster. Owing to heredity,

to self-reproduction, every form of life tends to transform the ma-

terials in the environment capable of serving as food into copies of

itself. Self-reproduction continues in adult bodies which no longer

grow in size or in weight; it continues as long as life itself. Every

organism not only assimilates foods but also breaks down the assimi-

lated products and reconstructs them again. This continuous buildup

and breakdown of living matter is the metabolism.

Studies on metabolism have disclosed that most of the components

of the animal body are frequently renewed, as can be shown very

clearly with the aid of "tracers," such as isotopes of the elements which

compose the living bodies. For example, an animal can be fed a diet

in which certain of the atoms of the usual carbon (Ci2) have been

replaced by those of a radioactive carbon isotope, C^. The radio-

active carbon presently appears as a component of various tissues of

the body, in which it can be localized with the aid of both physical

and chemical methods. The Ci4 thus enters the living protoplasm,

but after a lapse of a certain time (variable for different tissues and

substances), the structures containing Ci4 are broken down, and the

isotope appears in the waste products of the organism. Similar experi-

ments have been done with other isotopes—"heavy" water, radioactive

sulphur, phosphorus, iodine, etc. The phenomenon of self-reproduc-

tion in living bodies is universal.

Adaptation. Every organism responds in definite ways to outside

stimuli; the ability of making such responses is referred to as the prop-

erty of irritability. Even the simplest unicellular animals and plants,

such as amoebae and flagellate algae, move towards food and away

from too intense heat or cold, and towards moderate light and away

from darkness or too intense light. Seeds of the higher and spores of

the lower plants respond by sprouting or germination to some condi-

tions and by remaining dormant to others.

A striking and fascinating fact is that many, although not all, re-

sponses of living beings to outside influences are adaptive, that is, pro-

mote health, survival, and reproduction of the reacting organisms.

Homeostasis 13

Consider, for example, the complex physiological processes which are

set in motion by wounds, injuries, or bone fractures, and which result

in healing and repair of these injuries. Invasion of the body by dis-

ease-producing bacteria, viruses, and other parasites calls forth re-

markable defense reactions which tend to localize, combat, and finally

extinguish the infections. Owing to the scarcity of oxygen at high

elevations in the mountains, man experiences difficulties in living and

working at high altitudes. The human body reacts, however, to the

high-altitude conditions by altering the composition of the blood and

by other physiological changes which facilitate securing enough oxy-

gen for respiration.

Homeostasis. In all these and in countless other instances we ob-

serve a remarkable "wisdom of the body." Environmental stimuli

evoke processes which maintain or restore the welfare of the body in

changed environments. The organism normally stands to its environ-

ment in a relation of adaptedness. When the environment changes,

this harmonious relation may be disturbed and must be restored by

homeostatic reactions on the part of the organism. Adaptedness to

the environment can, of course, be observed not only in man but in

all organisms, animal and plant, primitive and complex. The ubiquity

of homeostatic reactions in the living world has even led some biolo-

gists to view adaptation as some kind of mystical principle. Many

vitalists (see page 19) regarded the ability of the organism to respond

adaptively to external stimuli as an elemental property of life. The

great Lamarck, one of the pioneers of evolutionism, built his theory

on the assumption that organs are always strengthened by intensive

use and weakened by prolonged disuse (Chapter 4). It will be shown

below (Chapters 5 and 6) that the origin of adaptations can be under-

stood better on the basis of Darwin's theory of natural selection, with-

out invoking any alleged mystical properties of living matter. For

the time being it is sufficient to point out that, although the wide-

spread adaptedness of living beings to their environment justly ex-

cites our admiration, in some instances adaptive reactions are con-

spicuously lacking.

The sun-tanning reaction of the human skin protects the body from

sunburns. There is, however, no protecting reaction to shield the body

from harmful X-rays, radium rays, and similar radiations. In general,

pain is a warning signal produced by injuries or by invasions of the

body by poisons and parasites. Yet in the human species the biologi-

cally "normal" and essential function of childbirth is accompanied by

intense pain. Most harmful substances are repellent, but some, such

14 Nature and Origin of Life

as narcotics, produce pleasant sensations instead of pain. Man must

rely on food for his supply of numerous vitamins essential for health;

but many other animals, even some mammals, can manufacture some

of these vitamins for themselves.

There are some basic facts which point towards a solution of the

problem of adaptation. In general, every living species reacts adap-

tively to the external stimuli which occur frequently in the environ-

ments in which this species has evolved. Thus the human species had

for countless generations to deal with the danger of sunburn, but it

did not encounter X-ray burns until very recently. The painfulness

of childbirth is probably a disharmony resulting from the compara-

tively recent acquisition by our species of the erect posture and from

the correlated changes in the pelvis bones. When the normal food

supply of a species contains an abundance of a certain vitamin, there

is no obvious advantage for this species to manufacture the vitamins

in its own body. Homeostatic reactions are not to be taken for granted

as gifts of nature; they must be understood and explained. The mod-

ern theory of evolution seems to provide a reasonable explanation.

Viruses—the Simplest Organisms. Invention of the microscope re-

sulted during the eighteenth and nineteenth centuries in the discovery

of a hitherto unknown world of tiny creatures. Some of them, par-

ticularly bacteria, appeared at first to be mere droplets of a viscous

substance. Yet, despite their small size, bacteria possess an immensely

complex organization, at any rate immeasurably more complex than

anything in inorganic nature. Then, in 1892, Ivanovsky discovered

still smaller organisms which are too small to be seen in ordinary light

microscopes. They were later named filterable viruses, since they are

small enough to pass through filters of unglazed porcelain and other

materials which retain ordinary bacteria. In recent years many vi-

ruses have been made visible by means of electron microscopes. In

1935 Stanley isolated from tobacco plants infected with the so-called

mosaic disease a protein which can form crystal-like structures (Figure

1.5). A minute amount of this protein injected into healthy plants

causes them later to develop the mosaic disease. Most important of

all, an amount of the virus protein about one million times greater

than the amount injected can be obtained from the diseased plants

several days after the injection.

Viruses known at present are a diversified collection of forms. Their

common property, apart from their small size, is that all of them are

parasites which develop only in living cells of animal or plant hosts.

In other words, no free-living viruses are known. Some viruses cause

16 Nature and Origin of Life

apart from its nucleoproteinaceous structure, is that the virus repro-

duces itself at the expense of materials in the tobacco leaves. To be

sure, the objection has been raised that the virus may somehow be

synthesized by the cells of the tobacco plant or may be preformed in

these cells. Indeed, the virus multiplies only in living, not in dead,

cells of tobacco leaves. The objection is not valid; many a parasite,

the living nature of which cannot be doubted, cannot be cultivated

outside the body of a host. Tobacco cells assuredly contain sub-

stances which are susceptible of being converted into the virus sub-

stance; otherwise the virus could not infect the tobacco plants. But

the conversion does not take place unless a particle of the virus is first

introduced into the plant. Similarly, human intestine contains sub-

stances from which intestinal parasites can be built, and grocery store

shelves contain materials which can be converted into human bodies.

Yet in all these cases the conversion occurs only in the self-reproducing,

living, bodies of a given species.

The Continuity of Life. The idea that life is, in the last analysis,

but self-reproduction of living matter is quite foreign to a mind un-

tutored in biological science. Until a few centuries ago nobody

doubted abiogenesis or spontaneous generation, that is, origin of living

creatures from non-living matter. Aristotle (384-322 B.C.) thought

that worms and snails were products of putrefaction and that plants

could arise also without seed. This hypothesis was accepted by such

pioneers of science as Bacon (1561-1626) and William Harvey (1579-

1657). Paracelsus (1493-1541) revealed the recipe, which, he said,

had until then been kept in greatest secrecy, for producing a "homun-

culus" or an artificial man. He might be produced, allegedly, through

putrefaction of the human semen in a cucurbit placed for 40 weeks

inside the belly of a horse. The power of heredity was taken equally

lightly. Plutarch (A.D. 46-125) records that human beings were born

of mares, asses, and goats. The notion that wheat and barley trans-

form into wild oats and other weeds was discussed in ancient Rome

by Virgil and Pliny, and was believed by some until about the seven-

teenth century. It has quite recently been resurrected by Lysenko

and his followers in Russia.

The continuity of life and heredity was revealed only gradually.

Redi (1626-1698) showed that fly maggots did not arise from putrid

meat. He placed decaying meat in a jar covered with fine gauze, so

that the flies were unable to reach the meat. No maggots appeared.

Spallanzani (1729-1799) found that meat or meat juice did not pu-

trefy unless spores of putrefaction bacteria were introduced with dust

18 Nature and Origin of Life

ous practical applications—pasteurization and sterilization so impor-

tant in canning and dairy industry as well as in medicine.

Hypotheses Regarding Origin of Life on Earth. If life comes only

from life, then every living creature which exists now is a direct de-

scendant of the first bit of living protoplasm which appeared on earth

(unless life arose from inanimate nature repeatedly). The origin of

the first life of necessity is a highly speculative issue. Indeed, our

inability to observe spontaneous generation in nature or to bring it

about artificially in laboratory experiments shows that life must have

arisen under some conditions which no longer obtain at present and

about which we can make only the vaguest guesses.

Attempts have even been made to avoid the issue by supposing that

life was introduced on our planet from other heavenly bodies. Spores

or other germs of life may have arrived on earth in a meteor or with

cosmic dust. But we are not certain that life exists or ever existed

outside the earth, and, if it did, could stand the transport through the

interplanetary space. In any case, we must face the problem of the

origin of life in the universe.

All life that now exists, including the simplest viruses, has as its

physical abode highly complex organic compounds, nucleoproteins.

Not only nucleoproteins but even their constituent amino acids and

nucleic acids are synthesized exclusively in living organisms and never

spontaneously from inorganic substances (some of the amino acids

can, however, be made synthetically in laboratories). Spontaneous

formation of these energy-rich compounds is highly improbable on

the basis of physicochemical considerations. Nevertheless, several

scientists, Oparin in Russia, Dauvillier and Desguin in France, Bernal

in England, Urey, Miller, and Blum in America, have tried to visualize

conditions under which chemical substances now formed only in living

organisms could have arisen without intervention of life.

In the early stages of the history of the earth the prevalent high

temperatures permitted the existence of water only in the form of

superheated steam. The atmosphere contained no free oxygen, but it

may have contained some simple carbon compounds, such as methane

(CH4), which are now found in the atmospheres of some stars. Simi-

larly nitrogenous compounds, such as ammonia (NH3), were probably

formed. With the cooling of the earth and the formation of liquid

water some chemical reactions became possible which are not likely

to occur now. This is particularly so because high-energy ultraviolet

radiations were then presumably reaching the earth's surface. They

are now absorbed in the upper atmosphere by oxygen and ozone.

Vitalism and Mechanism 19

Under such conditions simple aldehydes, such as formaldehyde

(CH2O), would be formed, and these might be slowly condensed into

sugars. Organic acids also could be formed and, reacting with am-

monia, could give other compounds, including simple amino acids,

such as glycine (CH2NH2COOH). Proteins would arise by linking

together several amino acids.

A possible reason why these chemical compounds do not now exist

in non-living media is that they would be used up as food by living

organisms soon after being produced. While life was absent, organic

compounds could accumulate in the water of lakes and seas. Even so,

spontaneous formation of complex proteins and nucleoproteins is a

most improbable event, at least in terms of short time intervals. Given

eons of time, a highly improbable event may, however, take place

somewhere in the universe. Such a "lucky hit" happened to occur on

a small planet, earth, a mere speck in the vast cosmic spaces. As soon

as a particle appeared that was able to reproduce itself, that is, to

engender synthesis of its copies from materials present in its environ-

ment, the evolution of life was launched. The primordial living par-

ticle may, for example, be visualized as a kind of simple virus. Of

course this virus must have been able to multiply in the inorganic

environment, or in environments containing only the organic com-

pounds accumulated through processes of the kind outlined above.

There are several reasons why the self-reproduction of particles is

stressed as the essential step with which life commenced. Self-repro-

duction of necessity implies growth through assimilation, maintenance

of definite organization, and transmission of heredity. It will be

shown in the following chapters that self-reproduction and heredity

may lead, through action of natural selection, to adaptation to the

environment and to progressive evolution. Formation of the first self-

reproducing particle, whatever might have been its precise chemical

makeup, was, at least potentially, the dawn of organic evolution.

Although at present self-reproduction is not known to occur except

in nucleoproteins, self-reproducing units of other composition might

have occurred on earth or in other parts of the cosmos.

Vitalism and Mechanism. Our understanding of the fundamental

life phenomena is admittedly sketchy, and the hypotheses about the

origin of life are only conjectures set up to promote further thinking

and experimentation. Science in general and biological science in

particular have only begun to explore the design and the working of

nature. Most biologists believe that the best working hypothesis is

that life phenomena involve merely complex patterns of interaction

20 Nature and Origin of Life

of physical forces and of chemical reactions. This assumption is called

mechanism. Its alternative, vitalism, is that, in addition to the forces

similar in kind to those operating in inanimate nature, life involves

powers which are restricted to the living world.

It is curious to think that vitalism was accepted by most biologists

in the seventeenth and eighteenth centuries, together with the belief

in the frequent occurrence of spontaneous generation (see page 16).

At that time life was regarded as a manifestation of a special vital force

(vis vitalis). Such a possibility could not be dismissed outright.

Physics knows several forms of energy: mechanical energy, heat, light,

electricity, and the energy of atomic nuclei. There may exist also a

vital energy, it is argued. The development of biology in the nine-

teenth and twentieth centuries has failed, however, to reveal the oper-

ations of such an energy, although some of the physical and chemical

processes involved in many biological phenomena, such as animal and

plant metabolism, muscle contraction, nerve conduction, and other

phenomena, have become partly known. This certainly does not

mean that all biological processes are now understood in physical and

chemical terms, but it is true that the realm of phenomena in which

manifestations of the vital force may still be suspected has been shrink-

ing steadily for at least two centuries. Few or no biologists now con-

sider the assumption of vital force a profitable hypothesis, and this

notwithstanding the downfall of the belief in spontaneous generation,

which seemed to increase the gap between the living and the non-

living.

Vitalism, however, continues to exist in a subtly changed form and

as a minority opinion. It is admitted that life is basically a complex

of physical and chemical processes. And yet Sinnott believes that

living cells possess a "psyche," a drive akin to human volition, which

presides upon the development of a fertilized egg into an embryo and

into an adult. The so-called finalist school, particularly in France

and in Germany, believes that the evolution of life has been governed

by a perfectionist urge, or by striving to produce man (cf. Chapter 14).

It is hard to see what is gained by such speculation. Just what is

the meaning of the assertion that the cell or the body is directed by a

psyche or by a perfectionist urge? If these factors somehow direct

the flow of physical and chemical processes, they must themselves be

forms of physical energy, even though peculiar and as yet unknown

ones. But this is a return to the old belief in a vital force, although

under different names. If, on the other hand, the psyche or the super-

natural guide is outside and above the physicochemical matrix of life

Suggestions for Further Reading 21

processes, a biologist is still obliged to find mechanistic explanations

of these processes. Consider, for example, the origin of life from inert

matter. Was it the psyche which built the first nucleoprotein molecule

from simpler compounds? • If so, the psyche must be some enzyme

or some energy-rich radiation. No matter how small is the push which

you want the supernatural to deliver to a material process, the super-

natural is inexorably debased to the level of a physical energy.

Acceptance of mechanism as a biological theory is not, however,

inconsistent with esthetic and religious views of existence. Man is a

complex of chemical compounds, but this fact alone is said to describe

him adequately only to those who would use him for fertilizer. Re-

gardless of what he is chenvcally, man yearns for sympathy and under-

standing. A primeval forest, a mountain meadow, or a coral reef

have beauty which is appreciated the more we become familiar with

the organisms which occur there. Religion leads us to appreciate the

meaning and value of the world and of man.

Suggestions for Further Reading

Gamov, G. 1952. The Creation of the Universe. Viking Press, New York.

This is a very well-written account of cosmic evolution, understandable to non-

physicists.

Bernal, J. D. 1951. The Physical Basis of Life. Routledge & Kegan Paul, Lon-

don.

Blum, H. F. 1951. Time's Arrow and Evolution. Princeton University Press,

Princeton.

Oparin, A. I. 1953. Origin of Life. 2nd Edition. Dover Publications, New

York.

The books of Bernal, Blum, and Oparin contain, among other things, interesting

speculations concerning the problem of the origin of life.

Plunkett, C. R. 1944. The primary physicochemical basis of organic evolution.

In J. Alexander's Colloid Chemistry, Volume 5, pp. 1173-1197. Reinhold, New

York.

An excellent discussion of the importance of self-reproduction as a basis of life

and of evolution.

Russell, E. S. 1945. The Directedness of Organic Activities. Cambridge Univer-

sity Press, London.

Sinnott, E. W. 1950. Cell and Psyche. The Biology of Purpose. University of

North Carolina Press, Chapel Hill.

The books of Russell and Sinnott outline modern variants of vitalistic theories of

life.

22 Nature and Origin of Life

Simpson, G. G. 1949. The Meaning of Evolution. Yale University Press, New

Haven.

Also available in a paper-covered edition. A general discussion of evolution and

its philosophical implications by an outstanding paleontologist.

Zirkle, C. 1936. Animals impregnated by the wind. Isis, Volume 25, pp. 95-130.

Zirkle, C. 1951. The knowledge of heredity before 1900. In L. C. Dunn's

Genetics in the Twentieth Century, Macmillan, New York.

The two papers by Zirkle give a review of the amazing and amusing specula-

tions concerning generation and heredity put forward from antiquity to relatively

recent times.

2

The Gene as the Basis

of Evolution

Every species of organism reproduces itself. The process of heredity

converts the food derived from the environment into more or less faith-

ful copies of the assimilating organism and of its parents and more

remote ancestors. The reproductive potentials of many organisms are

immense. With abundant food and at favorable temperature, the

colon bacteria, Escherichia coli, double in number about every twenty

minutes.

Heredity is a conservative force. If children and parents were com-

pletely identical, evolution could not occur. Heredity, however, is

opposed by a process of change, variability. Self-reproduction occa-

sionally results in an imperfect copy of the parental living unit, and

the altered copy, called a mutant, then reproduces the altered structure

until new mutations intervene. One of the major achievements of

biology during the current century has been the demonstration that

the units both of heredity and of mutation are bodies of molecular

dimensions, called genes.

In the last analysis evolution is a sequence of changes in the genes.

Darwin and other pioneers of evolutionary biology realized very clearly

the basic importance of heredity for understanding evolution. How-

ever, it is only during the current century, and particularly during the

last twenty to thirty years, that a theory of evolution based on the

findings of the study of heredity, genetics, has become possible. A

brief outline of the fundamentals of genetics is, therefore, essential in

a book dealing with evolution. Such an outline is given in Chapters

2, 3, 4, and also elsewhere in this book in connection with the various

problems of general evolutionary biology discussed.

23

Mendel's Law of Segregation 25

of a person was thought to be an alloy in which the heredities of each

of its four grandparents were represented by one-quarter, of each of

the eight great-grandparents by one-eighth, etc. (This was called

Galton's law of ancestral heredity.) Even so perspicacious a mind as

Charles Darwin's accepted these beliefs of his time. People un-

familiar with biology accept the notion of "blood" heredity even now;

indeed, th's notion is a part of the everyday language.

Mendel's classical experiments consisted in crossing different va-

rieties of garden peas and observing the distribution of the character-

istics of the parental varieties among the hybrid offspring. He pub-

lished the results of his work in 1865 in a single paper. Although he

stated very clearly the laws of heredity now regarded of great im-

portance, his work failed to attract the attention of contemporaries.

In 1900 these laws were rediscovered independently by Correns in

Germany, De Vries in Holland, and Tschermak in Austria. Their

importance was recognized, and the development of a new science,

genetics or the study of heredity, proceeded apace.

Mendel crossed varieties of peas differing in single contrasting traits,

or characters, such as a variety with purple and one with white

flowers, with yellow and with green seeds, with round and with

wrinkled seeds, tall and dwarf varieties, etc. In the first generation of

hybrids, called the Fi generation, the trait of one of the parents was

usually dominant, whereas the alternative trait of the other parent was

suppressed, or recessive. Thus the cross purple X white flowers gave

hybrids with purple flowers, yellow X green seeds gave dominance of

yellow, round X wrinkled seeds resulted in recessivity of wrinkled,

and tall X dwarf variety gave hybrids in which the tallness (domi-

nant) masked the dwarf size (recessive).

The FI hybrids were allowed to produce offspring. The progenies

of the F1 hybrids are the second generation (F2) hybrid progenies.

In these F2 progenies Mendel observed a fundamental fact: the domi-

nant and the recessive grandparental traits reappeared in about three-

quarters and one-quarter of the progeny respectively (a ratio 3 domi-

nant : 1 recessive). Thus Mendel obtained F2 offspring among which

approximately 75 per cent of the plants had purple and 25 per cent

white flowers (Figure 2.2).

Mendel's Law of Segregation. Segregation of the traits of the

parental varieties in the F2 generation of hybrids shows that these

traits are not inherited through miscible "bloods." They are trans-

mitted from parents to offspring by discrete bodies, which, many years

after Mendel's death, Johanssen has named genes. In the hybrids the

26

The Gene as the Basis of Evolution

P

generation

generation

generation

Figure 2.2. Mendel's law of segregation. Cross of a purple-flowered and a

white-flowered strain of peas, showing dominance of the purple color and segre-

gation of purple and white in the second generation of hybrids (F2). R stands

for the gene for purple, and r for the gene for white flower color. Black rings

symbolize purple-flowered and white rings white-flowered plants.

Mendefs Principle of Independent Assortment 27

genes neither mix nor contaminate each other. Thus the gene for

purple flowers does not fuse with the alternative (allelic) gene for

white flowers, even though the purple dominates the development of

the color in the flowers of the F, hybrid plants. The white flowers in

the Fj generation are just as white as in the original white variety.

They do not become pink or rose-colored, although the contrasting

alleles for the purple and white colors are carried for a whole genera-

tion side by side in the bodies of the FI hybrids. When the hybrid, or

heterozygous, plants form their sex cells the purple- and white-produc-

ing genes segregate. The sex cells are "pure": they carry either a

purple or a white allele. This is the law of segregation, or Mendel's

first law.

It is convenient to symbolize the dominant and the recessive gene

alleles by capital and by small letters respectively (Figure 2.2). Thus

the purple-flowering variety of peas can be written RR, and the white-

flowering, rr. The sex cells carry either R or r. The FI hybrid plants

are heterozygotes, Rr. In the process of formation of the sex cells by

Rr heterozygotes the alleles segregate, and equal numbers of R-con-

taining and r-bearing sex cells are produced. In the formation of the

Fz progenies these sex cells combine at random: R female cells or r

ovules are fertilized by R or r male cells in proportion to their abun-

dance

It can now be predicted that among the F2 plants approximately

one-quarter will be homozygous for R, that is, will carry two similar

alleles, RR (Figure 2.2). About half of the plants will be heterozygous,

Rr, and about one-quarter will be homozygous, rr. Because of the

dominance of R and the recessiveness of r, the homozygotes, RR, and

the heterozygotes, Rr, will be similar in having purple flowers, whereas

the homozygotes, rr, will have white flowers (Figure 2.2). Although

the RR and the Rr plants are not distinguishable in their flower color,

Mendel saw a way to test the validity of the prediction based on his

theory. He permitted the purple-flowered F2 plants to produce further

progenies (F3). About a third of these progenies had only purple

flowers; these were the progenies of RR plants. In about two-thirds

of the progenies purple- and white-flowering plants appeared in ratios

approaching 3 purple : 1 white; these were the progenies of Rr plants.

Mendel's Principle of Independent Assortment. In some of the

crosses made by Mendel the varieties of peas differed in two or more

conspicuous traits. For example, he crossed a variety with yellow and

round seeds with one having green and wrinkled seeds (Figure 2.3).

Mendel knew that the yellow color is dominant over green, and that

28

The Gene as the Basis of Evolution

p

generation

aabb

AaBb

generation

generation

Figure 2.3. Mendel's law of independent assortment. Strains of peas with

yellow and smooth seeds and with green and wrinkled seeds are crossed. The

letters A and a stand for the genes for the yellow and green colors, and B and b

for the smooth and wrinkled seed surfaces, respectively.

Mendelian Inheritance in Man 29

the wrinkled seed surface is recessive to round. The Ft generation

hybrid seeds were, as expected, yellow and round. What, however,

will the F2 hybrids be like? The traits of the seed color give in F2 a

segregation in a ratio of 3 yellow (dominant) : 1 green (recessive)

seeds. The seed surface segregates in a similar ratio, 3 round : 1

wrinkled F2 hybrid seeds. However, will the yellow color be linked

in segregation with the smooth surface, and the green color with the

wrinkled surface? Or will the characteristics of color and surface seg-

regate independently? Mendel showed that the latter possibility is

realized. The proportions of yellow and green colors among the

smooth seeds are the same as among the wrinkled ones. In other

words, the F2 generation consists of about nine-sixteenths (% X %)

of yellow smooth, three-sixteenths (% X Vi) of yellow wrinkled, three-

sixteenths (% X !4) of green smooth, and one-sixteenth (% X /4)

of green wrinkled seeds. This, then, is a segregation in the ratio of

9 double dominants : 3 with one dominant and one recessive : 3 with

the other dominant and the other recessive : 1 double recessive.

In terms of genes these facts are interpreted as shown in Figure 2.3.

Let A stand for the dominant allele which produces the yellow color

and a for its recessive green alternative; B for the dominant smooth

surface, and b for the gene which gives rise to a wrinkled seed surface.

The varieties crossed are AABB and aabb, respectively. The sex cells

of these varieties are AB and ab, and the FI is a double heterozygote,

AaBb. The essential point is that the allele pairs A-a and B-b assort

independently when the genes segregate during the formation of the

sex cells in the hybrids. The Fi hybrid forms, then, four kinds of sex

cells in equal numbers with the gene combinations AB, Ab, aB, and

ab. Random union of these sex cells gives the sixteen combinations

shown in Figure 2.3. Remembering that the gene A is dominant over

a, and B dominant over b, we can easily deduce that the different com-

binations of seed color and seed surface will appear in the ratio

9:3:3:1 indicated above.

The principle of independent assortment, or recombination, is some-

times referred to as Mendel's second law. Although this "law" is not

obeyed in some instances (see page 54), the facts which led to its

formulation are important. They show that the heredity transmitted

through sex cells is a mosaic of corpuscles, genes which are to some

extent independent of each other. The different variants, alleles, of

the same gene do not fuse but segregate in heterozygotes, and differ-

ent genes undergo segregation independently of one another.

Mendelian Inheritance in Man. Mendel's laws apply to all living

beings and thus represent perhaps the most fundamental biological

30 The Gene as the Basis of Evolution

laws yet discovered. Segregation and recombination of hereditary

traits have been observed in most diverse animals and plants. It would

be difficult to count the number of species of organisms in which gene

heredity has been observed to occur.

Soon after 1900, examples of Mendelian inheritance were found in

man. Of course, controlled hybridization experiments cannot be ar-

ranged in man, but this drawback is compensated for by the abun-

dance of data on the distribution of various traits in human families

and populations. Let us consider as an example the inheritance of

the ability to taste the substance known as phenyl-thio-carbamide

(PTC, for short). Solutions of this substance have an intensely bitter

taste to about 70 per cent of Americans, but to about 30 per cent they

are almost tasteless. The numbers of tasters (those able to taste PTC

solutions) and non-tasters (those who find them without taste) in

808 families with 2043 children are summarized in Table 2.1. The

TABLE 2.1

INHERITANCE OF THE ABILITY TO TASTE PTC SOLUTIONS

(After Snyder.)

Number Children

of

Families Parents Tasters Non-tasters

86 Non-taster X Non-taster 5 (?) 218 («)

(ttXtt)

289 Taster X Non-taster 483 (Tt) 278 (tt)

(TT or Tt X «)

425 Taster X Taster 929 (TT or Tt) 130 (tt)

(TT or Tt X TT or Tt)

explanation which accounts best for the data is that the tasters carry

a dominant gene T, and the non-tasters are homozygous for its reces-

sive allele, t. Marriages between two non-tasters (tt X tt) should

produce only non-taster children. Actually 218 non-tasters and 5

tasters have been recorded. These "exceptions" are due to the fact

that some people who are actually "tasters" do not find weak solu-

tions of PTC particularly bitter; in any large material on human hered-

ity the occurrence of some illegitimate children must also be reckoned

with. Some of the parents recorded as tasters were doubtless homo-

zygous (TT), and others heterozygous (Tt), for the gene T. Some

families in which one or both parents were tasters produced, accord-

ingly, only tasters, whereas others produced both taster and non-

taster children (Table 2.1).

Some Applications of Mendel's Laws 33

by observing Mendelian segregation in progenies of crosses in which

the parents differ in some traits. If all men were tasters, or all non-

tasters of PTC, we would not so much as suspect that the genes T-i

exist. As a matter of fact, the discovery that PTC tastes bitter to some

but not to other people was made by accident, when one chemist

working with this substance felt a discomfort which his non-taster

colleague found hard to understand. If albinos were unknown, or if

everybody in the world had brown eyes, or only blue eyes, the genes

for albinism and for the eye color would remain unknown.

A gene must undergo a change and be represented by at least two

alleles before its existence can be ascertained. Very rough estimates

of the total number of genes have nevertheless been attempted. These

estimates are of the order of 5000 to 15,000 genes in the sex cell of

the vinegar fly, Drosophila, and 10,000 to 100,000 genes in a human

sex cell.

Mendelian segregation and assortment are observed in the offspring

of crosses. Now, crossing or hybridization presupposes that the or-

ganisms reproduce sexually. In some lower organisms, especially

among bacteria, sexual reproduction was unknown until recently, and

the existence of genes in such organisms was not rigorously estab-

lished until Lederberg found that some strains of the colon bacteria

can be crossed. Recombination of traits has recently been found even

in bacteriophages, making it probable that these organisms, visible

only in electron microscopes, contain several genes. Only the simple

viruses such as the tobacco mosaic virus are perhaps comparable to

single genes. For this reason they are sometimes referred to as "naked

genes."

Some Applications of Mendel's Laws. As stated above, Mendel

crossed a variety of peas that had yellow and smooth seeds with a

variety that had green and wrinkled seeds. Owing to the recombina-

tion of genes he obtained in the F2 generation of hybrids not two but

four varieties: with (1) yellow smooth seeds; (2) yellow wrinkled

seeds; (3) green smooth seeds; and (4) green wrinkled seeds.

Crossing, hybridization, evidently may lead to production of new

varieties, which is important in the breeding of agricultural crops and

animals. Suppose that a variety of wheat is valuable because it is

genetically resistant to frost but has the drawback of being suscepti-

ble to rust fungi. Another variety has a satisfactory rust resistance

but is not frost resistant. It is then expedient to cross these varieties,

and to look among the F2 hybrids for plants resistant both to frost

and to rust. Of course plants devoid of either resistance are also

34 The Gene as the Basis of Evolution

likely to appear. Combinations of good as well as of bad genes are

impartially produced by the Mendelian segregation. It is, then, the

business of the breeder to pick out, to select, the valuable combina-

tions of genes. Some of the most valuable crop varieties have been

obtained by selection among hybrid progenies derived by crossing two

or several less valuable varieties.

When the parents crossed differ in more than two genes the diversity

of genetic constitution obtained by recombination may be very great.

With three genes eight homozygous gene combinations can be formed

among hybrids, with four genes sixteen, with five genes thirty-two, and,

in general, with n genes 2" gene combinations. In some F2 progenies

of parents differing in many genes no two individuals may appear

alike.

Gene Differences among Siblings. Segregation and assortment of

genes take place not only in the offspring of artificially made crosses.

In sexually reproducing organisms the individuals that mate differ

usually in many traits and in many genes. Every human being, tech-

nically speaking, is a hybrid, a heterozygote, for many pairs of alleles.

With 10 heterozygous genes there may be produced 210 (1024) kinds

of sex cells with different combinations of genes; with 20 genes,

1,048,576 kinds of sex cells; and with 250 genes, about as many kinds

as there are electrons and protons in the universe. Nothing can be

more certain than that only a negligibly small fraction of the poten-

tially possible gene combinations in any species are ever realized.

Nature is prodigal in the number of sex cells that are generated in

many organisms. In man a single ejaculation contains about 200 mil-

lion spermatozoa. Suppose that an individual who produces these

spermatozoa is heterozygous for thirty or more genes. It becomes

unlikely that any two spermatozoa will contain the same combination

of genes. Since most people are heterozygous for probably more

than thirty genes, it is most unlikely that any two spermatozoa, or any

two eggs, have the same genes. In other words, siblings, brothers and

sisters, rarely if ever receive the same complements of genes from

their parents. Only identical twins, who arise through division of a

single fertilized egg, carry the same genes. Unrelated persons would

differ, on the average, in more genes than do brothers or sisters. It is,

then, a reasonable guess that no two persons alive (identical twins

excepted) carry the same genes. Every human being is a carrier of

a unique, unprecedented, and probably unrepeatable gene complex.

This is true as well for sexually reproducing and cross-fertilizing spe-

cies other than mas.

Manifold, or Pleiotropic, Effects of Genes 35

Genetic Endowments of Parents and Children Compared. Accord-

ing to the "blood" theory, a parent transmits his heredity as a whole

to every one of his children. On this basis, if among your ancestors

there was a passenger on the Mayflower, you would possess a particle

of every one of his qualities. In contrast to this, the gene theory

shows that every parent transmits to his child only half of the genes

which he himself has. The father and the mother of a single child

have transmitted to the future generations one-half of their genes, but

the other half of the genes which they carry will be irretrievably lost.

However, every child gets a somewhat different set of genes from each

parent. A parent of two children has transmitted approximately three-

quarters of his genes, and failed to transmit one-quarter; a parent of

three children has handed down to posterity about seven-eighths, and

a parent of n children a fraction 1 — (%)" of his genes.

Looked at from the progeny point of view, siblings are likely to

have approximately half of their genes in common, whereas the other

half will be different. The "blood" theory regarded the heredity of

a child a fusion product of parental heredities. If this were so, the

heredities of brothers and sisters would be similar or even identical.

The outcome of hybridization, of sexual union of dissimilar parents,

had to be a leveling-off, a neutralization, dissolution, of the differences

between the heredities of the varieties crossed. The gene theory

leads to precisely opposite conclusions. Sexual reproduction continu-

ously generates new combinations of genes. The diversity of heredi-

tary constitutions is thus maintained and increased by sexual unions, a

fact that is of immense importance for evolution. In later chapters it

will be shown that sex may be regarded as an adaptation of living

matter which permits living matter to secure the evolutionary advan-

tages of gene combination.

Manifold, or Pleiotropic, Effects of Genes. In the examples of

Mendelian inheritance discussed above, a gene carried in the sex cells

was always spoken of as representing a discrete trait of the adult or-

ganism. A gene in peas caused a yellow seed color or a wrinkled seed

surface, and in man we found genes "for" tasting PTC, for eye color,

for amaurotic idiocy, etc. This way of describing and naming genes

is convenient because it is concise; but it is also misleading because it

seems to imply that every gene has an exclusive influence on one and

only one character. In reality, the gene theory does not assume that

the genes are rudiments or representatives of particular body parts or

of particular traits. The development of the organism is due to all

the genes acting together in concert. All the genes which the or-

36 The Gene as the Basis of Evolution

ganism has interact with the environment, and in so doing they make

the fertilized egg develop by stages into a fetus, an infant, a child,

an adolescent, an adult, an old man or an old woman, and finally a

cadaver. It should always be kept in mind that, despite the shorthand

designations which geneticists use in naming genes, many, and prob-

ably all, genes influence several or many traits of the organism which

carries them. Genes have manifold or pleiotropic effects.

Some early geneticists liked to speak of genes determining "unit

characters"; yet pleiotropism of genes was known already to Mendel.

He crossed a variety of peas with purple flowers, brown seeds, and a

dark spot on the axils of the leaves to a variety with white flowers, light

seeds, and no axillary spot on the leaves. In the segregation observed

in the hybrids, the just-mentioned traits of the flowers, seeds, and

leaves always stayed together as a unit. Their inheritance can be

accounted for by a single gene which visibly influences several traits.

The work of Morgan (1866-1945) has made the vinegar fly (Dro-

vophila melanogaster) and its relatives classical materials for genetic

studies. One of the variants (mutants) differs from normal (wild-

type) flies of this species by having vestigial wings (Figure 4.5). The

cross vestigial X normal gives an FI progeny with normal wings and

a segregation in a ratio 3 normal : 1 vestigial in the F2 generation.

The gene for vestigial (vg) is, accordingly, recessive to that for normal

wings (Vg). Careful comparison of vestigial and normal flies dis-

closes, however, that the wing size is by no means the only difference

between them. Vestigial flies have also the third joint of the halteres

(balancers) rudimentary, a certain pair of bristles on the body erect

instead of flat, and some of the reproductive organs changed in shape.

Moreover, vestigial flies deposit fewer eggs than normal, their life is

on the average shorter, and their larvae are the losers if they are made

to compete with normal larvae in crowded cultures. To a human

observer, the wing size is certainly the most striking difference be-

tween vestigial and normal flies. To the flies, the differences in

fecundity, longevity, and viability may be more important.

Examples of manifold effects of genes could be multiplied at will.

Studies on hereditary diseases in man and in higher animals have

shown that the genes produce not single traits but more or less com-

plex systems, or syndromes, of characters. These syndromes often in-

clude changes in many body parts, organ systems, and physiological

functions of the organism.

The origin of pleiotropism is easy to understand. The genes bring

about the development of the organism through physiological, and

Interaction of Genes in Development 37

ultimately biochemical, processes in the cells, tissues, organs, and the

whole body. The genes produce, or influence the production of,

enzymes which are so important in cell metabolism. It may be that

every gene is responsible for the production of one and only one

enzyme; on the other hand, it is possible that the same gene, at any

rate in higher organisms, makes different enzymes in different tissues

and at different stages of development in the same tissue. However

that may be, a change or a removal of an enzyme may alter pro-

foundly the metabolism of cells and of the body. An alteration of

this sort may result in a group, a syndrome, of changes.

"Characters" Are Abstractions. To think that genes determine

characters is misleading because a "character" or a "trait" is an abstrac-

tion which an observer makes to facilitate the description of his ob-

servations. For example, a manual of anatomy contains thousands of

names for parts, ridges, holes, and other structural details of bones,

muscles, and other organs. We may say that each name corresponds

to a "character," but the number of the structures named is limited

only by the convenience of those who have to talk and write about

these structures. Yet a critic of genetics argued that if we must as-

sume one or more genes for each structure, the number of genes will

be infinite, hence the genes do not exist!

The solution of this imaginary difficulty is simple enough. The or-

ganism with all its "characters" is an outcome of a process of develop-

ment, which is a system of physiological, and ultimately physico-

chemical events. In the last analysis, these events are by-products of

the self-reproduction of the genes. Characters or traits which we ob-

serve are the outward signs of the occurrence of the development

process. How many characters we observe depends on how we look

at the organism, on how careful and detailed our studies are, and

above all on how we choose to describe our observations. But the

genes are there, independent of how we talk or write about them.

Interaction of Genes in Development. Every gene may affect many

visible traits; most traits are influenced by several or by many genes.

Interaction of gene effects in development may lead to some ex-

tremely complex situations which are at first sight quite different from

simple Mendelian inheritance. Analysis of situations of this sort in

terms of the gene theory was one of the outstanding problems of

genetics during the first decades of its existence (roughly, from 1900

till 1920).

Breeds of domestic and laboratory animals show quite a variety of

coat colors. In horses, cattle, sheep, cats, dogs, rabbits, guinea pigs,

38 The Gene as the Basis of Evolution

mice, rats, poultry, pigeons, canary birds, and other species the coal

and plumage colors are determined by interaction of numerous genes.

We choose the inheritance of the coat colors in horses as an example,

although there is no agreement among the investigators about the in-

terpretation of some parts of this complex subject.

Coat Color in Horses. The entire coat of some horses, including

the mane, tail, muzzle, and lower parts of the legs, is chestnut (sorrel)

in color. Mating of chestnut stallions and mares gives only chestnut

foals. Chestnut is due to homozygosis for a recessive gene, b. The

dominant allele, B, of the gene turns chestnut into black. Black horses

are either homozygous, BB, or heterozygous, Bb, for this gene. An-

other dominant gene, 7, turns black into bay—a brownish coat with

black mane, tail, muzzle, and lower parts of legs. A black horse, then,

is always homozygous for the recessive allele, if, in addition to having

at least one dominant B (Figure 2.6). But neither 7 nor t has any

visible effects in the absence of B; accordingly, chestnut horses may

or may not carry 7. As a result, horses of these colors may have th(

following genetic constitutions:

Chestnut Black Bay

bb ii BB ii BB II

bb Ii Bb ii Bb II

bb II BB Ii

Bbli

The matings black X black produce, then, either only black or black

and chestnut foals. Black X chestnut may give black, chestnut, and

bay. All three colors may also appear in the matings bay X bay, bay

X black, and bay X chestnut. Addition of a further dominant, D,

transforms the coat into dun (dilute yellow), sand-colored ("buck-

skin'), or bluish ("mouse") coat, with black or chestnut mane, tail,

muzzle, lower parts of legs, and a stripe on the middle of the back.

Chestnut, black, and bay horses are, then, homozygous, dd, for the

recessive allele, d ( = non-dun). The dun color (BBHDD) is charac-

teristic of the wild horse of Mongolia (Equus przewalskii), which is

one of the progenitors of the domestic horse. A race of the same

species, known as tarpan, lived in the steppes of Eastern Europe until

it was finally destroyed a century ago (see Chapter 9). This race was

mouse colored and probably had the genes BBiiDD.

Two further genes, G and R, make the horse respectively gray and

roan. A foal homozygous or heterozygous for G (GG or Gg) is born

p

generation

BBII

Bay horse

bbii

Chestnut horse

generation

generation

Figure 2.6. Inheritance of some of the pelage colors in horses. Cross of a bay

and a chestnut (sorrel) horse gives bay in Fi, and bay, black, and chestnut in

the F2 generation. The bay color is produced by simultaneous presence of the

genes 7 and B; B without 7 gires black, whereas homozygosis for b yields chest-

nut regardless of whether 7 is or is not present.

39

40 The Gene as the Basis of Evolution

with only a few white hairs scattered in an otherwise chestnut, black,

bay, or dun coat (depending on the other genes). The proportion of

white hairs rapidly increases, however, with age, making the horse first

dappled gray and eventually white (with black skin). The gene for

roan (R) acts in a different way: it causes the appearance of rather

numerous white hairs interspersed with hairs of other colors, the

amount of white not greatly increasing with age.

There are several further genes the effects of which on the coat

color are not completely understood. A dominant gene causes the

horse to be pied with large white areas on the trunk (pied, piebald,

or pinto horses). The presence of white markings on the face and on

the lower parts of the legs seems, however, due to a quite different

gene, which is recessive to its "normal" allele, causing absence of white

markings on these parts. A "white" horse, having a light skin (that is,

not an old gray), may be due to extreme development of the "pinto"

markings. There is another gene, however, which gives a white coat

when homozygous and a characteristic pale ("palomino") coat in

heterozygous condition.

Polygenic Inheritance. Analysis of such traits as the coat color in

horses is facilitated by the fact that most of the genes which influence

the trait produce discrete and easily perceptible effects. Very often,

however, the trait is caused by interaction of several or many genes,

each of which taken separately produces only a small effect. Such

genes are referred to as multiple genes or polygenes.

Inheritance of the Skin Color in Man. Davenport has suggested

(1913) that the difference in skin color between Negroes and whites

may be due to the cooperation of two pairs of genes without domi-

nance. This view undoubtedly oversimplifies the situation. Consid-

erably more than two pairs of genes influence the skin color in man.

To explain the principle of polygenic inheritance, let us assume three

pairs of pigment genes, such that a Negro has the genetic structure

PiPiPzPzPaPa, and a white, Piptp^p^papa- Each gene denoted by a

capital letter causes production in the skin of a certain amount of

dark pigment. The effects of different genes are additive; that is,

each gene adds a certain amount of pigment regardless of presence or

absence of other pigment genes. A Negro, then, would carry six pig-

ment genes, and a white no pigment genes (this is admittedly inexact,

since whites who are not albinos always have some skin pigment).

A Negro X white marriage produces Fi hybrids (mulattoes), who

should be PipiPzp^PaPa- With three pigment genes the mulattoes

have, on the assumptions made, half as much skin pigment as the

Inheritance of the Skin Color in Man

41

Negro parent. A mulatto produces, then, eight kinds of sex cells, with

the genes PiP2P3, PiP&2, Pip2P3, piJVs, Pip2ps, pI^ps, pip2Ps, and

Pip2ps- Accordingly, marriages of mulattoes will give progenies of

the genetic constitution shown in Figure 2.7. It can be seen that

almost a third—20 out of 64 mulattoes—of the F2 generation will carry

three pigment genes and will accordingly have about the same skin

color as the mulattoes of the FI generation. Almost a half—30 out of

30

11

individuals

•— ISJ

00 *.

—

*

•5

1

•£

2 12

-

—

"• 6

n

;

n

ni

Number of genes for dark skin pigment

Figure 2.7. Inheritance of the skin color in the white X Negro cross, on the

assumption that the color difference is due to three pairs of pigment genes which

exhibit no dominance. The heights of the bars in the diagram show the per-

centages among the second generation hybrids of individuals with no, with one,

two, three, four, five, and six color-producing genes.

64—will have two or four pigment genes and will have somewhat

lighter or somewhat darker skins than the average Fi mulatto. About

six in 64 will have five pigment genes and will be intermediate be-

tween Negro and mulatto, and about the same number will have one

pigment gene and a skin intermediate between mulatto and white.

And only a minority—one out of 64—will have six pigment genes and

full Negro skin color or no pigment genes and a white skin color.

Marriages of mulattoes and Negroes or of mulattoes and whites pro-

duce children most of whom have skin colors intermediate between

those in the parents. Indeed, with three pairs of pigment genes at

play, three-eighths of the sex cells of the mulattoes would have two

pigment genes, three-eighths, one pigment gene, one-eighth, three

oigment genes, and one-eighth, no pigment genes (Figure 2.7). The

42 The Gene as the Basis of Evolution

mulatto X Negro cross would, then, give three-eighths of the children

with five pigment genes, three-eighths with four, and one-eighth each

with six and with three pigment genes. The mulatto X white cross

would give three-eighths of the progeny with two pigment genes,

three-eighths with one, and one-eighth each with three and with no

pigment genes. Marriages between mulattoes and either whites or

Negroes give mostly children intermediate between the parents in

their skin color.

"Blending" Inheritance. Polygenic inheritance was the last type

of hereditary transmission to be analyzed in terms of genes. The

more multiple genes participate in the formation of a trait, the smaller

are the visible effects of each separate gene, and the more difficult is

the analysis of the "blending" of the characters which is observed.

{The inheritance caused by polygenes certainly resembles what was

popularly believed to be heredity through "blood." As pointed out

above, the difference in the skin color between Negroes and whites is

caused by probably more than three pairs of genes. Different "whites"

vary in skin color—from albinotic to tan—and this variation quite apart

from the color variations due to skin exposure to sunlight. African

Negroes are likewise variable in skin color. It is probable that some

pigment genes are scattered in white populations and some genes for

lightness of the skin in Negro populations. Races differ mostly in

relative frequencies of genes in their populations (see Chapter 7).

Varieties and breeds of agricultural plants and animals and also

races and species of all organisms differ most often in traits caused by

polygenes. Classical Mendelism dealt chiefly with clearly alternative

traits, which can easily be described as present or absent in a given

individual. Most men are either tasters or non-tasters of PTC, either

albinos or non-albinos, either brachydactylous or with normal fingers,

etc. Polygenic traits are usually matters of "more or less" of some

quality. Such quantitative characters are described by measurement

or weighing, and studies of their inheritance require application of

often recondite statistical techniques, which have been evolved and

perfected especially in recent years.

The polygenes and the genes with discrete major effects are not

basically different kinds of genetic units. All conceivable intermediate

situations exist. Substitution of a gene for its allele may cause a drastic

change in the organism; or the change may be slight; or it may be so

minute that statistical techniques are needed for its detection. The

difference between major genes and polygenes lies not in their nature

but in the techniques used for their study.

Suggestions for Further Reading 43

Suggestions for Further Reading

Sinnott, E. W., Dunn, L. C., and Dobzhansky, Th. 1950. Principles of Genetics.

4th Edition. McGraw-Hill, New York.

This is one of the several available textbooks of general genetics, containing an

outline of the main facts and concepts of this science.

Mendel, G. J. 1866. Experiments on plant hybridization.

Translated from the original German by William Bateson. This paper appears

as an Appendix in the book by Sinnott, Dunn, and Dobzhansky, referred to above.

Mendel's paper is a true classic, and it should be read by every student of genetics

and of evolution as a splendid example of the application of scientific method In

modern experimental science.

Gates, R. R. 1946. Human Genetics. Macmillan, New York.

Neel, J. W., and Schull, W. J. 1954. Human Heredity. University of Chicago

Press, Chicago.

Sorsby, A. (Editor). 1953. Clinical Genetics. Butterworth, London.

Stern, C. 1949. Principles of Human Genetics. W. H. Freeman, San Francisco.

These four books are concerned specifically with the genetics of man, although

Stern's contains an excellent outline of general genetics as well. Sorsby's book, as

its name implies, is devoted chiefly to the description of the genetically conditioned

diseases. Gates' book is useful as an extended compilation and bibliography of

human genetics up to the time of the Second World War.

Mather, R. 1949. Biometrical Genetics. Dover Publications, New York.

Lerner, I. M. 1950. Population Genetics and Animal Improvement. Cambridge,

London.

The books by Mather and Lerner deal with the rapidly developing field of study

of quantitative or polygenic traits, and the mathematical and statistical methods

used in this study.

3

Chromosomes as

Gene Carriers

Almost 400 years ago the versatile philosopher Montaigne admitted

being completely baffled by the mystery of heredity. He thought that

he had inherited from his father a disease—a stone in the bladder; but

his father suffered from this disease some years after Montaigne was

born. How, then, could the father transmit to his son something which

he himself did not have at the time the son was conceived? And,

besides, the semen was believed to be mere liquid. How could a

liquid transmit a stone in the bladder?

To dispel even a part of the mystery which worried Montaigne

much biology had to be learned. At present we would say that Mon-

taigne did not inherit a stone in the bladder; what he inherited were

genes which engendered a constitution, a development pattern, which

included a predisposition towards formation of bladder stones. Fur-

thermore, we know that the genes have a physical basis in the chromo-

somes in the nuclei of the sex cells, which are highly organized struc-

tures with complex and orderly behavior that makes heredity possible.

The physical basis of heredity is necessarily also the physical basis of

evolution.

Sex Cells and Fertilization. Leonardo da Vinci (1452-1519), who

was so much ahead of his times in so many things, realized the basic

fact that the father and mother contribute equally to the heredity of

the child, as shown by the following quotation: "The black races of

Ethiopia are not the products of the sun: for if black gets black with

child in Scythia, the offspring is black. But if a black gets a white

woman with child, the offspring is gray. And this shows that the seed

of the mother has power in the fetus equally with that of the father."

From Aristotle on, until as late as the eighteenth century, most people

44

Nuclei and Chromosomes 45

believed that the mother furnishes inert matter and the father imparts

the motion to the new life. Spallanzani (1729-1799) found, how-

ever, that this "motion" was not due to some immaterial essence in

the seminal fluid of the male, as others supposed; he showed that the

seminal fluid of the frog lost its ability to fertilize the eggs after a

passage through a filter. The fertilizing agent is not a simple liquid.

We could not go much beyond this without making use of a new

instrument invented and gradually perfected a generation before Spal-

lanzani's—the microscope. Using microscopes, Leeuwenhoek, Swam-

merdam, and others discovered that the seminal fluid contains "ani-

malcules"—spermatozoa (see Chapter 10). At about the same time,

de Craaf and others found that female mammals produced eggs like

birds and frogs, only much smaller. But more than a century had to

elapse, and more powerful microscopes had to be manufactured, be-

fore Oskar Hertwig (1849-1922) finally, in 1876, saw the eggs of a

sea urchin being fertilized by spermatozoa, and was able to discern

that the most significant event in the process is the union of two about

equal nuclei, that of the egg and that of the sperm. Very soon there-

after Weismann, Roux, Hertwig himself, and others realized that the

phenomenon of the fusion of the nuclei during fertilization explains the

equal potency of the female and the male in the transmission of

heredity from parents to the progeny. The sex cells, eggs and sptjr-

matozoa, in most organisms are as strikingly unlike as cells can be. Yet

they contain similar parts—their nuclei and chromosomes. The infer-

ence was then clear: the material basis of heredity resides primarily

in the nuclei and their chromosomes.

The quarter of the century immediately preceding the rediscovery

of Mendel's laws, roughly from 1875 till 1900, saw a rapid develop-

ment of cytology, the study of the cell. The following quarter of a

century, approximately 1900-1925, brought an even more rapid prog-

ress of cytogenetics, a synthesis of the findings of cytology and genetics

concerning the mechanisms of the transmission of heredity.

Nuclei and Chromosomes. Cell nuclei were described by Brown in

1831, even before the promulgation of the cell theory by Schleiden

and Schwann in 1839. Soon thereafter (in 1848) Hofmeister observed

the process of division of living cells in the plant spiderwort (Trades-

cantia). He saw in the nuclei of dividing cells rod-like bodies, which

later were found to stain deeply in fixed preparations by certain dyes,

and called since 1888 chromosomes (stainable bodies). These stain-

ing reactions are now known to be due to the class of substances called

46 Chromosomes as Gene Carriers

nucleoproteins, and particularly desoxyribose nucleic acids (DNA),

which all chromosomes contain.

The number of chromosomes in a nucleus was found, with few

exceptions, to be constant for a species. In many organisms different

chromosomes in the cell nuclei are recognizably different from each

other. This led Boveri (1862-1915), Wilson (1856-1939), Navashin,

and others (around 1900) to the correct inference that chromosomes

are unlike in their genetic contents (Figures 3.1 and 3.2).

Strasburger, Biitschli, Fleming, Roux, and others found that cells

divide usually by a remarkably precise mechanism called mitosis (Fig-

ure 3.3). Mitosis begins with a prophase stage, when chromosomes

appear in the nucleus as slender threads, which are often composed

of bead-like chromomeres. The chromomeres may be of different sizes

and shapes, making each chromosome in a nucleus recognizable by a

definite sequence of large and small chromomeres following each other

in a longitudinal file. The constancy of the chromomere pattern is

visible evidence of the chromosome being longitudinally differentiated

into qualitatively different segments. This differentiation reflects the

constant linear arrangement of the genes (see page 58).

The chromomere structure of the chromosomes is particularly evi-

dent in the giant chromosomes of cells of the salivary glands of larvae

of certain flies (Figure 3.4). These giant chromosomes, first described

by Balbiani in 1881, and correctly interpreted by Heitz, Bauer, and

Painter in 1933, are used extensively in genetic and evolutionary

research.

The prophase stage is followed by metaphase. The chromosomes

shorten and thicken (by being thrown into a fine spiral). A spindle-

shaped figure, composed of thin fibers or threads, arises in the cyto-

plasm. The nuclear membrane disappears, and the chromosomes be-

come arranged, usually in a single plane, midway between the poles

of the spindle. The chromosomes now divide longitudinally into ex-

actly equal halves, and during the anaphase the halves pass to the

opposite poles of the spindle. The cell also divides, the chromosomes

enter the telophase and form the nuclei of the daughter cells (Figure

3.3).

In most organisms the disjunction of the daughter halves of the

chromosomes and their distribution to the poles of the mitotic spindle

are governed by the centromeres. A centromere is a specialized seg-

ment of the chromosome, permanently fixed in position, which seems

to act as the insertion point of the fiber connecting the chromosome

with the pole of the spindle.

Meiosis

51

of the gametes contains the sum of the chromosomes which were car-

ried in the gametes. The diploid chromosome numbers are"48-in man.

8 in Drosophila melanogaster, 20 in corn. ^

Figure 3.6. Diagrammatic representation of the stages of meiosis shown in Fig-

ure 3.5. The diagrams show only a single pair of chromosomes, the paternal

chromosome being represented black and the maternal white. The centromeres

are shown as white circles.

Sooner or later the diploid zygote must give rise to haploid gametes.

This change is accomplished by a remarkable modification of the

mitotic cell division, known as meiosis. In seme lower organisms

52 Chromosomes as Gene Carriers

meiosis occurs almost immediately after the formation of the zygote

by fertilization; in higher organisms the body consists of diploid cells,

and meiosis takes place in the sex glands or in the flower buds. How-

ever, the essential features of meiosis, like those of mitosis, are similar

in otherwise diverse organisms. We are evidently dealing here with

a fundamental life process.

The events which constitute meiosis may be seen in Figures 3.5 and

3.6. A diploid cell enters what resembles at first a mitotic prophase,

but at the pachytene (or zygotene) stage the corresponding (homolo-

gous) maternal and paternal chromosomes approach each other and

pair side by side. Where the chromosomes are differentiated into

chromomeres it can be seen that the pairing is very exact: the homolo-

gous chromomeres lie side by side. The attraction forces which bring

together the homologous chromomeres must be amazingly specific.

Every one of the many genes in the nucleus finds without fail its

proper partner. In any case, the chromosome pairing reduces the

diploid number of single chromosomes to a haploid number of paired

bivalents.

At the transition between the pachytene and the diplotene stages

(Figures 3.5 and 3.6) the chromosomes become visibly divided. The

bivalents consist now of two pairs of chromosome strands. In most,

although not in all, organisms the pairs of strands can be seen to be

held together by chiasmata. A chiasma involves an exchange, or

crossing over, of sections between the paired chromosomes, and pro-

duces chromosome strands composed of parts of the maternal and the

paternal chromosomes. Because of the chiasma formation the chromo-

somes are not inherited as units. Blocks of genes are exchanged be-

tween homologous chromosomes (see below).

After the diplotene, the chromosomes shorten (diakinesis), and the

nucleus undergoes two meiotic divisions during which no further di-

vision of the chromosomes takes place (Figures 3.5 and 3.6). Conse-

quently, each of the four nuclei resulting from meiosis contains a

haploid chromosome complement. The fate of the four nuclei is dif-

ferent in different cases. In the spermatogenesis of animals each

nucleus gives rise to a spermatozoon; in oogenesis three of the foui

nuclei are thrown out into "polar bodies," and only the fourth becomes

the nucleus of the egg; in higher plants meiosis gives rise to macro-

spores and microspores, which produce respectively the embryo sacs

with the ovules and the pollen grains. But in all cases the gametes

come to contain haploid sets of chromosomes.

54 Chromosomes as Gene Carriers

traits, which showed Mendelian inheritance in crosses with normal

flies and with each other. However, in some of these crosses certain

traits failed to show the expected independent assortment. Examples

of such crosses are shown in Figures 3.8 and 3.9.

When a fly with vestigial wings but with a normal, gray, body color

is crossed with a fly which has a black body but normal wings, the

FI hybrids are wild-type, that is, have normal wings and a normal

body color. Vestigial wings are, then, recessive to normal wings, and

black is recessive to the normal gray. Let these hybrids be back-

crossed to a double recessive strain with vestigial wings and black

bodies. According to Mendel's second law, the hybrids should pro-

duce four kinds of gametes in equal numbers, and the progeny of the

backcross should consist of black-vestigial, black, vestigial, and wild-

type flies in equal proportions (1:1:1:1).

This expectation is not realized. When FI hybrid males are back-

crossed to black vestigial females the offspring are vestigial gray-bodied

and black long-winged flies in equal numbers (Figure 3.8). When

FI hybrid females are used, the four expected classes appear in the off-

spring, but not in equal numbers. The parental combinations of

genes, that is, the vestigial gray and the black long-winged flies con-

stitute about 83 per cent of the progeny. The products of recombina-

tion of genes, wild-type and black vestigial flies, make up only about

17 per cent of the progeny.

Morgan saw that this infringement of Mendel's second law is ex-

pected if the genes for the black body color and for vestigial wings

are borne on the same chromosome. The FI hybrids shown in Fig-

ures 3.8 and 3.9 carry one chromosome with the normal allele of

vestigial and the gene for black. The genes in a chromosome are

linked in inheritance. At present more than 500 genes are known in

Drosophila melanogaster. They fall into four linkage groups, corre-

sponding to the 4 chromosomes in the haploid set of this species. In

maize ten linkage groups are known that correspond to the ten chromo-

somes in the haploid set. In the mouse thirteen linkage groups are

known; seven more linkage groups remain to be discovered, since the

haploid set consists of 20 chromosomes.

Linear Arrangement of Genes. The linkage of genes located in a

chromosome need not be absolute. In the vestigial X black cross 17

per cent of recombination appeared in the offspring of FI hybrid

females (Figure 3.9). The occurrence of chiasmata (page 52), of

the exchanges of sections between the paired paternal and maternal

Sex Chromosomes 55

chromosomes at meiosis, is responsible for this recombination of linked

genes. Morgan advanced the hypothesis that the frequency of recom-

bination is a function of the distance between the genes in the chromo-

some. Other things being equal, the farther apart the genes are, the

more likely it is that a chiasma will be formed between them. Neigh-

boring genes in a chromosome are strongly linked, those farther apart

are exchanged more frequently. Of course the likelihood of chiasma

formation depends on many things besides the distance between the

genes. It happens that in the spermatogenesis of male Drosophila

no chiasmata are formed in the chromosomes. Accordingly, no re-

combination of linked genes takes place in the offspring of hybrid

Drosophila males.

Morgan's hypothesis was amply confirmed and developed by Bridges,

Muller, Sturtevant, and others. Numerous experiments were carried

out in which the recombination of linked genes was studied in hybrids

of Drosophila and in other organisms. Sturtevant found a regularity

which holds strictly for rather closely linked genes: namely, if the

frequency of recombination between genes A and B is x, and between

B and C is y, then the recombination between A and C is either

x + y or x — y. Such a relationship is expected if the genes A, B, and

C are arranged in the chromosome in a single linear file. Although

complications arise with loosely linked genes, the experimental data as

a whole are consistent with the theory.

This theory has enabled geneticists to map the distribution of genes

in the chromosomes of genetically well-known species. Figure 3.10

shows maps of this sort for the most thoroughly studied form, Dro-

sophila melanogaster. These "genetic" maps indicate which genes

belong to each linkage group; the order in which the genes of a given

linkage group are arranged in the chromosome; and the "distances"

between the genes, expressed in "map units," which, in turn, represent

the frequencies of recombination between the genes expressed in

percentages.

Genetic chromosome maps, all of them less detailed than those

shown in Figure 3.10, now exist for several species of Drosophila, for

maize, for peas (Pisum), sweet peas (Lathyrus), beans, morning glory,

the fungus Neurospora, the mouse. Very sketchy maps, showing the

location of very few genes, exist for some chromosomes of about two

dozen more species of plants and animals, including man.

Sex Chromosomes. During the early years of the current century,

Wilson, Sutton, McClung, Montgomery, Stevens, and others investi-

58

Chromosomes as Gene Carriers

vidual will have two X-chromosomes, and will be a female. But if a

spermatozoon without an X fertilizes an egg, the result is an odd num-

ber of chromosomes, a single X-chromosome, and a male.

j,s

1

yellow body

1 5.- : 7-white eyes

r

1

0.0-,

..aristaless antenna

-Star eyes

0.0-

-•-rougtioid eyes

3.0-;',- ./facet eyes

u-"

5.5'',.- -..'ichinus eyes

7.5' "ruby eyes

13.7- - crossveintess wings

I3.fr-

-dumpy wings

20.0-.. ,.-cut wings

ttS-

clot eyes

m

--javelin bristles

21.0 - singed bristles

26.0.,

..sepia eyes

27.7 -lozenge eyes

26.5-:

- -hairy body

33.0 -vermilion eyes

361 miniature wings

41.0.

.Dichaele bristles

432 "

'..thread arista

44.0 -garnet eyes

44.0-;

..scarlet eyes

48.5

black body

48.a ,!

...oink eyes

—curled wings

50.0-

56.7. .forked bristles

57.0 :v ::-:• Bar eyes

59.5 fused veins

54.5-.

54.8-J

55.0?

.purple eyes

N Bristle

Vlight eyes

58.2

58.5N

58.7--=

Stubble bristles

/spineless bristles

'- bithorax body

52.5 carnation eyes

66.fr— L- bobbed hairs

57.5

67.0-

tinnabar eyes

vestigial wings

60 Chromosomes as Gene Carriers

ing a Y. All the eggs have 24 chromosomes, one of which is the

X-chromosome. Fertilization restores the diploid chromosome comple-

ment, 48; but in about half the cases this complement will contain two

X-chromosomes, and will give a girl. The other half of the fertilized

eggs will carry an X- and a Y-chromosome, and will give a boy.

Curiously enough, in birds, butterflies, moths, some fish, and prob-

ably in at least some of the amphibians and reptiles, the conditions

are reversed. Namely, it is the male which has two X-chromosomes

and forms spermatozoa all of which carry an X. Females are XY,

and half of the eggs carry an X, whereas the other half carry the

Y-chromosome.

Sex-Linked Inheritance. In 1910 Morgan found a mutant indi-

vidual of Drosophila melanogaster which had white instead of the

normal red eyes. This fly was the progenitor of a strain of white-eyed

flies. When white-eyed females from this strain were crossed to nor-

mal red-eyed males, the offspring consisted of red-eyed daughters

and white-eyed sons, as shown in Figure 3.12. The F2 generation had

about equal numbers of white-eyed and red-eyed females and males.

The reciprocal cross, normal, red-eyed females to white-eyed males,

gave the result shown in Figure 3.13. The FI flies were red-eyed,

whereas in FZ half of the males were white-eyed, and the other half

of the males and all the females were red-eyed.

This sex-linked inheritance follows if the dominant gene for red,

and its recessive allele for white, eyes are borne in the X-chromosomes.

The Y-chromosome has no allele of this gene. A male derives his

single X-chromosome from his mother; a female receives one X from

her mother and another X from her father. A father transmits his

X-chromosome to all of his daughters but to none of his sons. The

Y-chromosome is transmitted from father to sons only in the male line.

An interesting historical detail is that in 1910 Morgan's hypothesis

did not appear to be free of difficulties. Proposing this hypothesis

required a great deal of courage. Indeed, it so happened that, before

the studies of Drosophila, sex-linked inheritance was discovered in a

species of moth. But moths, as indicated above, have a sex-determin-

ing mechanism which is the reverse of that in Drosophila, XX males

and XY females. Accordingly the sex-linked inheritance in moths is

also reversed: a female transmits her sex-linked genes to all of her

sons but to none of her daughters. In 1910 the sex chromosomes of

Drosophila and of moths had not been studied cytologically, and all

the insects which had been examined had XY males (or XO males,

devoid of Y-chromosomes). The inheritance of genes and the inheri-

tance of chromosomes did not seem to agree very well.

Changes in Chromosome Numbers 63

increased if the parents are treated with X-rays or temperature shocks.

In 1916 Bridges (1889-1938) published his classical work analyzing

these exceptions.

Bridges reasoned that a white-eyed daughter of a red-eyed father

must possess two X-chromosomes (since she is a female), both carry-

ing the gene white (since she has white eyes). She can have re-

ceived these X-chromosomes only from her mother. A red-eyed son

must have a single X-chromosome (since he is a male), derived from

his father (since he has red eyes). How can such distribution of the

sex chromosomes take place? Bridges's hypothesis was that the process

of meiosis in the females goes wrong in about one among 2000 to 3000

cells (Figure 3.14). Instead of the two X-chromosomes of the female

disjoining normally, both of them either remain in the egg or are

eliminated in the polar body. This failure of the chromosome dis-

junction yields "exceptional" eggs with two X-chromosomes or with

no X-chromosome.

An egg with two X's may be fertilized by a Y-bearing spermatozoon

and will give rise to a white-eyed female. This exceptional female

must, if the hypothesis is correct, differ from normal females by having

a Y-chromosome in addition to her two X-chromosomes. Normally,

of course, the Y-chromosome is found only in males. Here, then, is an

opportunity to test the validity of the hypothesis. Bridges proved

cytologically that the exceptional females have the extra Y-chromo-

some as predicted. An exceptional egg with no X-chromosome, fer-

tilized by an X-bearing spermatozoon, will give a red-eyed "excep-

tional" male. The exceptional males, however, must lack the Y-

chromosome. This prediction also proved correct.

The exceptional eggs with two X's, fertilized by an X-bearing sperm,

will give individuals with three X-chromosomes. Such individuals are

poorly viable, but they occasionally survive as so-called superfemales.

They were identified by Bridges in another experiment. The eggs

devoid of X-chromosome, fertilized by a Y-bearing sperm, are iuviable;

such dying eggs were identified in the experiments of Li and of Poul-

son. Bridges analyzed also the progeny of the exceptional, XXY,

females crossed to normal males; he was able to make certain pre-

dictions regarding the composition of these progenies which were

found to be correct.

Changes in Chromosome Numbers. Occasional failures of the nor-

mally very precise processes of mitosis and meiosis may have important

consequences. Non-disjunction gives rise to cells with one of the

chromosomes represented more, or fewer, times than the other chromo-

Changes in Gene Dosage

65

the cell division may fail, despite the chromosomes' having normally

split into daughter halves. Such failures of the cell division occa-

sionally occur in apparently normal individuals. Within the last two

decades chromosome doubling without cell division has been induced

artificially with the aid of certain drugs, particularly the alkaloid

abed

e hi a b c f a d e f g hi

abed e f g hi a b c d

e f ff hi

c.- .ixyit. > dS3^nzD

a b c d e f ff hi
a. b

ff f e
Figure 3.15. A scheme showing different kinds of chromosomal aberrations. A.

Two pairs of original or "normal" chromosomes. B. Deficiency. C. Duplication.

D and F. Heterozygous and homozygous translocations, respectively. E and G.

Heterozygous and homozygous inversions, respectively. (From Sinnott, Dunn, and

Dobzhansky, courtesy of the McGraw-Hill Book Company.)

colchicine. As a result of the chromosome doubling, sex cells arise

which carry a diploid instead of the normal reduced, or haploid,

chromosome set. Whole individuals are thus obtained with three

(triploid), four (tetraploid), or higher polyploid chromosome com-

plements. Polyploidy is of considerable importance in the evolution

of the plant kingdom (Chapter 9).

Changes in Gene Dosage. Most genes are represented once and

only once in the haploid chromosome complement. However, from

time to time individuals appear which have a block of genes lost (de-

ficiency) or present in excess (duplication) (Figure 3.15). Just as

66 Chromosomes as Gene Carriers

with aneuploids, deficiencies and duplications for large sections of

chromosomes are often inviable. It is interesting that the organism

withstands the duplication of a block of genes generally more easily

than a deficiency of the same genes. In particular, the loss of a gene

or a group of genes from both chromosome sets in a diploid organism

(a homozygous deficiency) is usually lethal. This fact strongly sug-

gests that almost every gene which the organism has must be present

at least once to permit life and development.

Deficiencies, and especially duplications, have doubtless played

important roles in evolution. The evolutionary development of the

living world has, on the whole, led from simple to more complex forms

of life. It is reasonable to suppose that this progression from the

simple to the complex was accompanied by an increase in the number

of genes which a species carries. Duplication and polyploidy are the

only known methods whereby such increase could occur, since the

appearance of self-reproducing genes from non-self-reproducing cell

structures seems improbable. When a duplication occurs, the genes

in the repeated sections are, to start with, merely copies of each other.

In the process of evolution, however, they may suffer divergent

changes (by mutation) and thus become different genes.

Changes in Gene Arrangement. The linear arrangement of genes

in the chromosomes is usually constant in all individuals of a species.

However, changes in the chromosome structure occasionally occur;

their frequency is materially increased in the offspring of X-ray-treated

individuals. The starting point of these changes is breakage of the

chromosomes. Suppose, for example, that two different normal chromo-

somes, which carry the genes ABCD and EFGH, respectively, break

into fragments AB, CD, EF, and GH. The fragments are usually lost

unless the broken-off ends re-establish connections with other broken-

off ends. New connections may, however, arise thus: ABGH and

EFCD. Such exchange of segments between different chromosomes

is known as translocation (Figure 3.15).

A single chromosome may be broken at two or more points. Thus

the chromosome ABCD may give fragments A, BC, and D. The

middle fragment may rotate through 180 degrees, and a new chromo-

some may have a section inverted compared with the original ar-

rangement: ACBD. This is known as inversion of a chromosome

section (Figure 3.15).

There are some simple rules which govern the formation of trans-

locations and inversions. In organisms in which the chromosomes

have localized centromeres, each rearranged chromosome must have

Chromosomal Aberrations 67

one and only one centromere. Chromosome fragments devoid of

centromere become lost. Similarly, new chromosomes formed by the

union of fragments are lost if they include two centromeres. Cyto-

logical examination of dividing cells exposed to strong doses of X-rays

usually shows chromosome fragments which are eliminated from the

nuclei on account of failure to include a centromere. Such cells come

to contain deficiencies for blocks of genes and eventually die off. This

effect of X-rays on chromosome breakage is responsible for a major

part of the radiation damage to living tissues, as well as for the re-

gression of X-ray-treated cancerous growths.

Genetic and Cytological Study of Chromosomal Aberrations. Com-

parative genetic and cytological investigation of deficiencies, duplica-

tions, translocations, and inversions has yielded final proof of the

validity of the theory of linear arrangement of the genes. Every one

of these chromosomal aberrations can be diagnosed in genetically well-

known organisms by making crosses with strains containing suitable

genetic "markers." And in organisms favorable for cytological inves-

tigations the aberrant structure of the chromosomes can be seen under

the microscope.

Consider, for example, the study of the Notch deficiency in the fly,

Drosophila melanogaster, made by Mohr (1923). Females heterozy-

gous for Notch have a notched wing margin, and there are certain

disturbances in the arrangement of the small bristles on the thorax of

the fly. Males which receive the X-chromosome containing Notch

from their mothers are inviable. Notch is, accordingly, a sex-linked

condition which has a dominant effect on the wings and bristles when

heterozygous, and a recessive lethal effect in the male.

Mohr crossed Notch females to males with white eyes. As we know,

white eyes in Drosophila are due to a recessive sex-linked gene (Fig-

ure 3.12). As expected, half of the females in the FI generation of

this cross showed Notch wings; the other half had normal wings. But,

unexpectedly, the FI Notch females also had white eyes, whereas the

non-Notch females had normal red eyes. In the presence of Notch

the white eye color behaves as though it were dominant to the nor-

mal red. Another normally recessive trait, namely, facet eye, also

acted as a dominant in Notch flies, but other recessive sex-linked traits

behaved quite normally (see the "map" of the X-chromosome in Fig-

ure 3.10). Mohr concluded that Notch is not due to a change in a

single gene, but represents a deficiency in the chromosome comprising

a block of genes which includes white and facet.

Under the microscope a deficiency can be detected most easily in

Evolution of Heredity 69

or homozygous for deficiencies or duplications are usually different

from normal in external appearance, and sometimes inviable. But

translocations and inversions involve changes only in the arrangement

of genes in the chromosomes. Translocation and inversion heterozy-

gotes and homozygotes should have the same genes as individuals

free of these chromosomal aberrations, even though these genes are

differently arranged in the chromosomes.

It might seem that the appearance of the organism and its physio-

logical functions should not be changed by the occurrence of trans-

locations or inversions. This is, indeed, often the case. But many

exceptions are known, particularly in Drosophila. In some instances

normally dominant genes lose their dominance when placed in the

chromosomes with changed gene order. In other instances the re-

arrangement of the genes results in some of these genes' behaving as

though they had suddenly become very unstable and had undergone

frequent changes, or mutations, during the development of the or-

ganism. This last type of behavior gives rise to "spotted," or "mosaic,"

distribution of colors and other traits, not uncommon among garden

varieties of some ornamental plants. Finally, some translocations and

inversions are poorly viable or lethal when homozygous.

The above position effects are forcing geneticists to revise their ideas

about the relationships between the genes and chromosomes. Until

the discovery of the position effects there was nothing to contradict

the assumption that a chromosome is an aggregate of completely inde-

pendent units, genes, arranged in a fortuitous linear order. This was

certainly the simplest hypothesis that we could make, and it served

well for a time; but the real situation is not quite so simple. The genes

which lie in the chromosome next to each other are neighbors because

their proximity makes them act together well. A chromosome is not

just a container for genes but a harmonious system of interacting genes.

The arrangement of genes in a chromosome has developed gradually

during the evolution of the organism to which the chromosome be-

longs; the structure of a chromosome, like the structure of any organ,

is a product of adaptive evolution.

Evolution of Heredity. Among the now-existing organisms possibly

only the simplest viruses may be simple self-reproducing molecules,

or "naked genes." All other organisms have many genes. It is tempt-

ing to speculate that life appeared at first in the form of virus-like

molecules which caused formation of their copies from non-living

substances in the environment. These primordial viruses then became

diversified in response to the variety of environments which could

70 Chromosomes as Gene Carriers

sustain their self-reproduction (Chapter 5). The next step in the

progressive evolution involved association of two or several unlike

simple viruses into compounds; development of a mutual usefulness

and interdependence; and final integration of the associated viruses,

or genes, into organismic units, which might have been primitive cells.

Cells again formed colonies. At first the colonies consisted of equiva-

lent and later of differentiated members. The latter became multi-

cellular organisms.

Association of interdependent self-reproducing bodies, or genes,

raised biological problems which the "naked genes" did not have to

face. Consider what happens when a complex of interdependent

units, such as a cell, gives rise to a progeny. The progeny must be

endowed with a fixed number of copies of each of the constituent

units. The reproduction of compound units must be accurate. In the

process of evolution this problem of precision was solved by assembling

genes into aggregates, known as chromosomes, and elaboration of

mechanisms of cell division, or mitosis.

The sexual process brought with it the evolutionary advantage of

continuous production of ever-new gene combinations (Chapter 11).

Sexual reproduction, however, requires mechanisms that would alter-

nately place together gene sets derived from different parents and

recombine and sort out new gene complements. This requirement

became satisfied with the appearance and perfection of the mech-

anisms of fertilization and of reduction division, or meiosis. Aggre-

gation and interdependence of genes in complex organisms thus led

to the evolution of the genes being supplemented by the evolution of

chromosomes.

Suggestions for Further Reading

Sinnott, E. W., Dunn, L. C., and Dobzhansky, Th. 1950. Principles of Genetics.

4th Edition. McGraw-Hill, New York.

Chapters VII-XI of this textbook of genetics discuss the basic facts of cyto-

genetics.

White, M. J. D. 1954. Animal Cytology and Evolution. 2nd Edition. Cam-

bridge University Press, London.

This is the best available outline of cytogenetics of animals and of its bearing

on evolutionary problems.

Riley, H. P. 1948. Genetics and Cytogenetics. John Wiley, New York.

A textbook of genetics with an emphasis on plant cytogenetics.

Suggestions for Further Reading 71

Wilson, E. B. 1928. The Cell in Development and Heredity. 3rd Edition.

Macmillan, New York.

Though antiquated, this is a classic worth reading to anybody interested in

genetics or cytogenetics.

Morgan, T. H. 1919. The Physical Basis of Heredity. Lippincott, Philadelphia.

This is another classic which played a tremendously important role in the cor-

relation of genetic and cytological findings.

Heredity, Environment,

and Mutation

Everyday language uses the word "heredity" both for biological in-

heritance and for legal inheritance of property. The statement that a

person has inherited from his parents a dark, or a light, skin color does

not mean the same thing as the statement that this person has in-

herited from his relatives a house or a farm. Inherited houses and

farms are buildings or pieces of land which change their owners; the

skin color is not transferred in this manner. The narrow bridge which

connects parents and offspring is formed by the sex cells, and human

sex cells have no color, and for that matter no skin. Inheritance of

skin color refers to a developmental pattern which leads to the for-

mation of a certain amount of pigment in the skin.

Just how much skin pigment is formed depends, however, not only

on the presence of certain genes but also on the environment in which

the carriers of these genes live. Exposure of the skin to sunlight, or to

ultraviolet rays of certain wavelengths, darkens the skin color. The

trait (skin color) that develops is thus determined by interaction of

the heredity and the environment.

The environment concerned, however, is not only that prevailing

at a given moment, but also the whole sequence of environments which

the organism met during its lifetime. The color of my skin today

is determined not only by the sunshine or its absence now but also

by the amount of time spent outdoors and indoors during the preceding

months. The personality of every human being is determined by his

heredity, upbringing, education, relationships to other persons, disease,

nutrition, etc. Every one of us is a product of his life experience, his

biography. Nobody can escape his past. Living organisms are time-

binding machines; they are products of their histories.

72

Genotype and Phenotype 73

Heredity in Different Environments. All organisms, from simplest

viruses to man, build their bodies, and those of their offspring, from

materials derived from the environment. Every organism can re-

Droduce itself from a certain range of food materials, and under a

variety of environmental conditions. An organism that could exist in

only a single environment would not remain alive for long, because

the environment does not remain the same from one instant to the

next. Every organism, therefore, is adapted to live in a certain variety

of environments. How can this adaptedness be retained in changing

environments?

Let A stand for a self-reproducing entity— a gene, a virus, or the

sum total of the genes which an organism carries (the genotype, see

below). Suppose that A builds its replicas in different environments

or from different materials (foods) BI, B2, B3, etc. The process of self-

reproduction may, then, be visualized as follows:

A + BI =
A + B2 = 2A + C3

Ci, C2, and C3 may stand for by-products of the gene reproduction,

or for the non-self-reproducing cell parts, or for the bodily forms which

are the outcomes of the development of the organism in different en-

vironments. The essential point of this scheme is that, so long as the

environment is capable of maintaining life of a given kind of gene or

of gene system, the genes reproduce their copies ( A becomes 2A ) , or

else they do not reproduce at all. In contrast to this, the organisms

(C) produced in different environments vary.

Genotype and Phenotype. The fundamental nature of the distinc-

tion between A and C in the above scheme was perceived in 1909 by

Johannsen (1857-1927). He called the heredity received by an or-

ganism its genotype, and the appearance of this organism its pheno-

type. The phenotype changes continuously so long as the organism

remains alive. A series of photographs, taken at intervals from infancy

to old age, illustrates the changes in the phenotype of a person.

On the other hand, the genotype of a person is relatively constant.

It is presumably much the same during manhood as it was in youth

and in childhood and as it will be during senescence. The nature of

this constancy must be clearly understood. Although the chemistry

of the gene reproduction is unknown, it is quite certain that synthesis

of nucleoproteins must entail a complex series of chemical reactions.

74 Heredity, Environment, and Mutation

Far from being inert and insulated from the environment, the genes

are perhaps the most active cell constituents. The constancy of the

genes is, then, singularly dynamic—they interact with the environment

to transform a part of it into their own copies.

Germplasm and Soma. Even before Johannsen, Weismann (1834-

1914) drew the distinction between the germplasm and the soma. The

germplasm has its seat in the sex cells and the cells of the reproductive

organs which give rise to the sex cells. The soma is the rest of the

body. The germplasm is potentially immortal; indeed, every sex cell

is able, under favorable conditions, to give rise to a new individual

with another crop of sex cells. The soma is mortal; it is the body which

houses the sex cells, and which is cast off in every generation owing

to death.

Weismann's concepts of germplasm and soma were an important

landmark in the process of understanding heredity and evolution. But

they should not be confused with the modern genotype-phenotype

concepts. Not all the genes are carried in the reproductive cells; they

are present as well in every body cell, in other words also in Weis-

mann's soma.

Norm of Reaction. The question whether the genotype or the

environment is more important in the formation of the phenotype

or the personality is evidently meaningless (although frequently and

acrimoniously discussed). The phenotype is the outcome of a process

of organic development. There is no organic development wi'.hout an

organism, and no organism without a genotype. Equally, every or-

ganism exists in an environment and at the expense of an environment.

As pointed out above, any organism is the product of its genotype

and of its life experience or biography.

The genotype determines the course which the development may

take in any environment. In other words, the genotype of an individual

determines the norm of reaction of the individual in all possible

environments (Figure 4.1). A newborn infant has a great range of

possible futures; which of these possibilities are realized and become

actualities depends upon the succession of the environments gradually

unfolding during the lifetime. The ranges of possibilities are, how-

ever, different for carriers of different genotypes. For example, the

genotype of an albino causes little or no pigment to form regardless

of the amount of skin exposure to sunlight. The genotypes of most

"whites" make the skin colors vary within wide limits, depending upon

exposure to sunlight. And the genotypes of Negroes make the skin

76 Heredity, Environment, and Mutation

ments in ways that enable it to secure its livelihood. The survival

depends upon the norm of reaction.

In Chapter 1 (pages 13-14) it has been pointed out that many, al-

though not all, modifications of the phenotype produced by changes

in the environment are adaptive. The organism reacts to many en-

vironmental changes by homeostatic modifications of the developmen-

tal patterns, which favor the perpetuation of life in the changed

environments.

However, different genotypes often differ in their ability to produce

adaptive responses to environmental changes. The albinos in man do

not develop the protective sun tan, and may suffer dangerous sun-

burns. The albino genotype is not well adapted to environments with

intense sunshine. The genotypes of most "whites" develop the pro-

tective tan if their carriers are exposed to sunshine, but the pigment

is lost after a prolonged lack of sun exposure. This last reaction is

also believed by some authorities to be adaptive, since unpigmented

skin may facilitate the formation of vitamin D (the "sunshine vita-

min") in climates and during seasons when sunshine is scarce. The

genotypes of "whites" are, then, adaptive in seasonally changeable

climates, such as those of northern and central Europe. It is no

accident that most human races which inhabit the tropical and sub-

tropical regions have dark skins. These are the regions of abundant

and intense sunshine (see Chapter 13).

Superior and Inferior Norms of Reaction. When carriers of a

genotype are placed in an environment to which their genotype is not

adapted they react by loss of health or by death. Hereditary defects

and diseases are genotypic variants which react to environments usual

for the species or race by production of ill-adapted phenotypes. There

is, consequently, no hard and fast distinction between "hereditary"

and "non-hereditary" diseases. Most or all human beings may come

down with measles if exposed as children to an environment contain-

ing the virus of measles. On the other hand, only some people develop

diabetes. Accordingly, we call measles a non-hereditary and diabetes

a hereditary disease. But there may be some persons genotypically

immune to measles. And a person with hereditary diabetes may enjoy

good health if he receives injections of proper amounts of insulin at

regular intervals. Either genotype may produce disease in some

and health in other environments.

We should beware of assuming that some genotypes are always

"normal" and others "abnormal," except in the sense that some are met

with more and others less often in nature. The more frequent geno-

Non-inheritance of Acquired Characters 77

types, by and large, are those which are better fit to survive under

usual, frequently met with, environmental conditions. But it may be

misleading to say that the carriers of a certain genotype must

reach certain "intrinsic" height, or weight, or skin color, or intelligence

level. Any height or weight or intelligence a person may have is

"intrinsic," in the sense that the phenotype observed is the necessary

outcome of the development brought about by a certain genotype in

a certain succession of environments. We can never be sure that any

of these traits have reached the maximal development possible with

a given genotype. The performance of a genotype cannot be tested

in all possible environments, because the latter are infinitely variable.

Non-inheritance of Acquired Characters. It is a matter of every-

day experience that children resemble their parents; the influence on

organisms of nutrition, climate, and living conditions is also evident.

Putting these facts together, popular imagination concluded that en-

vironmental modifications of the bodies of parents are transmitted by

heredity to the offspring. The belief that acquired characters are

inherited was originally not a scientific theory; it was, and still is, a

part of the folklore.

Buffon (1707-1788) was one of the precursors of evolutionism, who

believed that the environment may change the nature of organisms

("denature" them, as he preferred to put it). He took it for granted

that the changes induced by the environment in what we would now

call the phenotype would be inherited. Lamarck (1744-1829), the

first thorough-going evolutionist, emphasized particularly the fact that,

in animals, extensive use or exercise of organs strengthens them,

whereas continued disuse weakens them. For example, muscles be-

come larger and stronger as a result of exercise, and are reduced by

prolonged disuse. It seemed evident to Lamarck that such acquired

changes would be transmitted to the offspring. Darwin (1809-1882)

considered natural selection (Chapter 6) the fountainhead of evolu-

tion. Nevertheless, he accepted the inheritance of acquired modifica-

tions as an important, if subsidiary, force, a view that was shared

by most of Darwin's immediate followers. For example, the Negro

race was considered "a child of the African sun." The tanning of the

skin by exposure to intense sunlight over many generations was be-

lieved to produce a progressive darkening, which becomes fixed in the

heredity of the inhabitants of Africa. Similarly, the absence of eyes

in many subterranean animals was ascribed to inheritance of the

effects of long-continued disuse of the eyes, owing to the life in

darkness.

78 Heredity, Environment, and Mutation

The assumption that acquired characters are inherited was finally

challenged by Weismann (whose principal work was published in

1892). Weismann's famous experiment consisted in cutting off the

tails of newborn mice in a series of successive generations. The tails

were no shorter in the progeny of experimental mice. This experi-

ment seems rather naive at present, but it is only fair to say that

heritable degeneration of an organ as a result of its amputation in

the parents is even now an accepted belief among some of the fol-

lowers of Lamarck.

Many experiments were carried out in the closing decade of the

nineteenth, and the first two decades of the current century, to test the

hypothesis of inheritance of acquired traits. The results of these ex-

periments were overwhelmingly negative. Most biologists came to

the conclusion that the hypothesis must be rejected.

One of the last alleged instances of such inheritance created some-

thing of a stir, because it was espoused by the eminent physiologist

Pavlov (1849-1936) in Russia and McDougall in America. In Mc-

Dougall's experiments rats were dropped into a water tank with two

exits, one lighted and the other dark. The rats had to learn to use the

dark exit, the lighted one being provided with a mechanism which

gave an electric shock to an animal attempting to use it. Different

rats require different numbers of "lessons" to be taught. It was con-

tended that the offspring of trained rats were taught more easily than

those of the untrained ones, apparently an inheritance of acquired

training. Repetition of these experiments disclosed a more complex

situation than was originally suspected. Among rats there exist strains

which differ genotypically in their response to training. The original

experiments involved an unconscious selection of rats with norms of

reaction favorable to training as parents of the next generations.

The Hypothesis of Pangenesis. The view that the earth is round

was not accepted without struggle, because everybody could so easily

see that the earth is flat. Scientific theories gain acceptance with diffi-

culty if they contradict popular beliefs. The negative outcome of ex-

periments on the inheritance of acquired traits failed to convince

some people. Is it possible that the outcome of such experiments

would be different if the action of the environment inducing the traits

were continued ten times, or a thousand times, longer than it actually

was? Perhaps the most satisfactory answer to such doubts is that the

mechanism of heredity revealed by modern genetics makes inheritance

of acquired characters highly improbable.

How, indeed, can the skin color produced by sun exposure influence

The Hypothesis of Pangenesis 79

the genes in the sex cells which will determine the skin color of the

next generation? An imaginative solution of this problem was sug-

gested by Maupertuis (1698-1759), made popular by one of Darwin's

bitterest critics, Samuel Butler, in 1879, and worked out in detail

by the vitalist Semon (1904). They compared biological heredity

to memory. The sex cells, as it were, "remember" the structure of

the body which produced them, and choose materials from the en-

vironment which can build a similar body. The acquired characters

are simply "remembered." The trouble is that this ingenious theory

fails to explain anything. Ascribing memory to sex cells, like ascrib-

ing a vital force to the living matter, merely takes for granted the

natural phenomena which have to be explained.

Darwin, who, as we know, also believed in the inheritance of ac-

quired characters, proposed in 1868 his "provisional hypothesis of

pangenesis." He assumed that all organs of the body, perhaps all

cells, produce diminutive vestiges of themselves, called gemmules or

pangenes. The gemmules are shed into the blood stream, and are

transported by blood to the sex glands, where the gemmules of dif-

ferent organs are assembled to form sex cells. Body cells modified

by the environment might produce modified gemmules which will re-

produce the modification in the next generation (see Figure 4.2).

The hypothesis of pangenesis had the virtue of being easy to test.

Galton in 1875, and other investigators later, made experiments of

blood transfusion, or of transplantation of ovaries, between white and

black varieties of rabbits and of poultry. The hypothesis would lead

us to expect the birth of spotted, black and white, offspring from

experimental animals. This expectation was not fulfilled.

The hypotheses of pangenesis, and of heredity as "organic memory,"

are invalid. The reason why they are discussed at all is that, as pointed

out by Zirkle (1946), variants of these hypotheses are, explicitly or

implicitly, a necessary part of every belief in the inheritance of ac-

quired traits. The Russian agriculturist Lysenko has gained much

notoriety owing to the suppression of genetics in the USSR. Lysenko

is a believer in the inheritance of acquired traits, and, apparently un-

aware of Darwin's authorship, he propounds an hypothesis of pan-

genesis as his own invention. Similarly, he ascribes a power of mem-

ory to the sex cells.

In reality, the genes do not arise from gemmules cast off by the

body cell; they reproduce by synthesizing their own copies. The self-

reproduction of the genes sets the norm of reaction of the organism in

different environments.

82 Heredity, Environment, and Mutation

Darwin considered "sports" too rare to be of much importance in

evolution. He ascribed more importance to the small, "fluctuating,"

inheritable differences which occur between individuals of any species.

Men, even members of the same family, differ in height, shades of

skin, hair, and eye colors, shape of the head and face, and many other

traits.

De Vries (1848-1935) took the opposite view; in his Mutation

Theory (1901) he maintained that evolution proceeds by large, dis-

crete, and sudden changes, which he called mutations. He supported

his thesis with observations on the evening primrose, Oenotltera la-

marckiana, which in his garden produced several mutations. De

Vries regarded mutations as new species of plants, and their real

nature was discovered only much later. Some of them represented

chromosomal aberrations: triploids, tetraploids, or aneuploids. Others

were homozygotes for recessive genes for which the parents were

heterozygous. Only a minority were due to mutational changes in

the genes which took place in the experimental plantings.

Is Evolution Continuous or Discontinuous? After the publication

of de Vries's work there was much discussion between Darwinists and

mutationists. Darwinists contended that evolution resulted from grad-

ual shifts in the characteristics of a species over a long series of

generations. Mutationists thought that evolution consists of rela-

tively rare but drastic mutations. At present, we know that this issue

was a spurious one.

Beginning in 1910, Morgan and his collaborators described many

mutations in species of vinegar flies, Drosophila. Some of the muta-

tions produce striking changes: flies with vestigial wings or with no

wings at all, with eyes of bizarre shapes or without eyes, with yellow

or with black bodies instead of the normal gray, etc. Some mutations

are so drastic that they kill the organism; such mutations are called

lethal. But many, in fact most, mutants produce changes so slight that

an expert eye is necessary to notice them at all. Mutations range all

the way from minute to drastic changes. Evolution is a gradual and

continuous process, but it results from the summation of many dis-

continuous changes, mutations, a great majority of which are small.

Mutationism is certainly not opposed, but supplementary, to Dar-

winism.

Mutations in Different Organisms. Mutations have been observed

in diverse organisms, from man to the simplest viruses. When the

mutants can be crossed to the ancestral form, the mutant traits are

usually inherited according to Mendel's laws. Most mutations appear

Homeotic Mutants 83

to be changes in single genes. The mutant alleles are sometimes

dominant ones, but more often recessive to the ancestral condition.

Some mutations, however, produce changes in the chromosome struc-

ture, chromosomal aberrations (Chapter 3).

Contrary to de Vries's opinion, mutations do not produce new

species. The mutants of Drosophila are still flies which belong to

the same species of Drosophila to which their ancestors belonged

(see, however, the species formation through polyploidy, Chapter 9).

Mutational changes affect all organs of the body and all kinds of

traits. Drosophila mutants differ in the color of the body, of the eyes,

and of some internal organs; in size and shape of the whole body, of

the eyes, wings, bristles, legs, eggs, larvae, and pupae (Figures 4.4 and

4.5). The viability, length of life, fertility, and behavior of the flies

may be altered. Physiological and biochemical traits may be affected.

Because of the work of Beadle and others, numerous biochemical

mutants have recently been found in the fungus Neurospora, in yeasts,

and in some bacteria. These mutants show a great variety of meta-

bolic patterns. They require the presence in their food of vitamins,

amino acids, or other substances, which are not needed for normal

growth of the ancestral form or of other mutants. The cells of the

mutants accumulate chemical substances which in the ancestral forms

are intermediate products of certain metabolic reaction chains. Such

biochemical mutants are invaluable for the study of cell physiology.

A new branch of science, biochemical genetics, is evolving on the bor-

derline between genetics and chemical physiology. Apart from its

intrinsic interest, this science promises also valuable technological

applications.

Lethal mutants cause the death of their carriers. Death may occur,

in different lethals, at any stage of the development, from cleavage of

the fertilized egg to the adult organism. Poulson, Hadorn, and others

have studied the causes of death in some lethals in Drosophila; and

Dunn, Landauer, and others in mice, poultry, and other animals. Some

of the lethal "syndromes" resemble certain human hereditary diseases.

Gene changes by mutation may drastically modify the basic develop-

mental processes of the organism.

Homeotic Mutants. Fundamental, as well as superficial, traits are

under the control of genes, and are subject to mutation. Of course

the distinction between the "fundamental" and the "superficial" is,

to a large extent, arbitrary. This is particularly clear in homeotic

mutants, which transform some organs into others. In at least three

species of Drosophila mutants are known which turn the "balancers"

Heredity, Environment, and Mutation

TABLE 4.1

MUTATIONS OBSERVED IN SEVEN GENES IN MAIZE

Gene

R (color factor)

7 (color inhibitor)

P-i (purple color)

Su (sugary)

Y (yellow seeds)

Sh (shrunken seeds)

Wx (waxy seeds)

(After Stadler.)

Individuals

Examined

Mutations

Observed

554 , 786

273

265,391

28

647,102

7

1,678,736

4

1,745,280

4

2,469,285

S

1,503,744

0

Mutations

per 1.000,000

Individuals

492

106

11

2.4

2.2

1.8

0

Mutations in any single gene may be rare. But since a chromosome

complement contains thousands or tens of thousands of genes (Chap-

ter 3), many mutations arise in most species in every generation. In

Drosophila melanogaster, approximately 2 out of every 1000 X-chro-

mosomes acquire a new lethal mutant gene per generation. Mutations

which produce small changes are doubtless more numerous than

lethals. The frequency of such small mutations is little known, since

it is hard to notice them all, although they are doubtless more frequent

than lethals.

Thus it comes about that Timofeeff-Ressovsky (1939) has estimated

that, in Drosophila melanogaster, 2 to 3 per cent of individuals in

every generation carry one or more newly arisen mutant genes. Mul-

ler's estimate for man is even higher. Ten to 40 per cent of human sex

cells in every generation carry a newly arisen mutation. The stability

of the gene as a self-reproducing unit is compatible with the produc-

tion of numerous mutants which serve as raw materials for evolu-

tionary changes.

Environment and Mutation. For more than half a century biologists

have tried to discover the causes which bring mutations about, and

the problem is still far from being solved. We do not know how to

produce a given kind of mutation where and when desired.

Mutations appear as a rule in single individuals so that only one of

the many sex cells produced by the parents of a mutant carries a gene

which suffers a given kind of mutational change. Mutation in a di-

ploid cell affects only one chromosome, although two chromosomes

of the same kind, with more or less similar genes, are present in the

cell. This isolated occurrence of mutations is suggestive. The imme-

Chemical and Biological Mutagens 87

diate causes, whatever they may be, which make the gene mutate are

strictly localized within a cell, in the immediate vicinity of the gene

which undergoes the change. It may be, although this is not proven,

that mutations occur when the genes reproduce themselves. One or

both copies of a gene formed by the self-reproduction process may

differ from the original in structure.

However that may be, several environmental factors are known

which speed up the mutation process. Muller, who was later awarded

a Nobel prize for his discovery, announced in 1927 that X-rays are

mutagenic (mutation inducing). Muller observed many mutations in

the progeny of X-rayed Drosophila. Depending upon the amount of

X-rays applied, mutations may be tens and even hundreds of times

more frequent in the progeny of irradiated flies than in the non-irradi-

ated control flies.

X-rays are mutagenic in all animals and plants which have been

experimented with, and there is every reason to think that they are

mutagenic in all organisms, including man. The fact that the number

of mutations induced is proportional to the amount of the rays reach-

ing the sex cells is important. X-rays are now used extensively in

medicine, and many people are becoming exposed to the radiations

produced by the release of atomic energy. Misuse of X-rays may raise

the specter of uncontrolled increase of the mutability in human popu-

lations, a disaster for public health.

Chemical and Biological Mutagens. Several strongly mutagenic

chemical compounds are now known. Experimenting with Drosoph-

ila, Auerbach (1949) found that mustard gas is a powerful mutagen.

It is one of the "war gases" (C1-CH2-CH2)2S. When applied in con-

centrations just short of those which kill the flies, this gas increases the

mutation rates almost as much as the strongest practicable X-ray

treatments. Several chemicals are mutagenic when mixed with the

food on which Drosophila larvae grow and develop. Among them,

the effects of formalin and urethane are best ascertained. Demerec

has discovered (1952) that the mutation rates in certain bacteria are

greatly increased by treatments with manganous and ferrous com-

pounds. Organic peroxides are also mutagenic, at least in bacteria.

An environmental factor that may be important as a mutagen under

natural conditions is high temperature. Muller and Timofeeff-Res-

sovsky (1935) found that, in Drosophila, mutation rates are doubled

or trebled with every 10°C rise in temperature. Although this is a

relatively small increase compared with what can be produced by

88 Heredity, Environment, and Mutation

X-ray treatments, it may be important for evolution if the mutation

rates remain high for many generations.

Doubtless important, and yet little studied, are the genetic modifiers

of the mutability. Ives (1950) found that different strains of the

same species of Drosophila show different mutability in the same en-

vironment. In some instances it was possible to show that such dif-

ferently mutable strains differed in a single gene, which did not have

any visible effects on the flies other than the modification of the mu-

tation rates. The mutability of a gene depends, therefore, not only

upon its own structure but also upon other genes which the organism

carries. The physiological mechanisms which bring about such modifi-

cations of the mutability are completely unknown.

Spontaneous and Induced Mutations. When mutants appear in the

progeny of parents which were not treated with any known mutagens,

the mutations are said to have arisen spontaneously. Of course, the

word "spontaneous" applied to any natural phenomenon means only

that the actual causes of this phenomenon are unknown. After the

discovery of the mutagenic effects of X-rays and similar radiations,

a suspicion arose that the spontaneous mutability may be caused by

cosmic rays and other high-energy radiations. The amounts of such

radiations, however, proved sufficient to account for less than one per

cent of spontaneous mutations.

It is important to remember that X-rays, and all other mutagens so

far discovered, merely increase the frequency of the same kinds of

mutations which also arise spontaneously. In other words, when we

apply a mutagen to a culture of Drosophila or of bacteria, the treat-

ment is likely to give in different individuals all sorts of mutations in

all genes, instead of definite mutations in genes of our choice. Al-

though the discovery of these mutagens has been a great achievement

of biological science, we cannot help wishing that means would be

found to transform genes in desired ways. In theory, there is no reason

why specific transforming principles could not be found. The search

for them has indeed met with some success.

Cytoplasmic Heredity. Apart from the genes carried in the chromo-

somes, some self-reproducing units exist also in the cytoplasm of some,

and perhaps of all, cells. Examples of such units are the chloroplasts,

the chlorophyll carriers of green plants. The chloroplasts arise from

self-reproducing primordia in the cytoplasm. In recent years evidence

has been accumulating that the cytoplasm of many organisms contains

plasmagenes, self-reproducing structures too small to be seen even

with the aid of microscopes.

Cytoplasmic Heredity 89

When a cell divides, the genes in the chromosomes are apportioned

to the daughter cells by means of the high-precision mechanism known

as mitosis (Chapter 3); but there is no mechanism of comparable ac-

curacy for the division of the plasmagenes. A cell usually carries sev-

eral or many plasmagenes of each kind, and at cell division each

daughter cell receives roughly half of the plasmagenes present in the

mother cell.

Sonneborn, Beale (1949), and others have observed remarkable

transformations in the infusorium, Paramecium aurelia. When infu-

soria of this species are injected into rabbits, the blood serum of the

rabbit comes to contain antibodies which paralyze the infusoria placed

in a drop of liquid containing this serum. Different strains of in-

fusoria, however, induce formation of different antibodies, so that the

infusoria can be classified into groups, or "serotypes," A, B, C, D, etc.

Under standard culture conditions, each strain of infusoria breeds true

as to its serotype. But the serotype can be changed.

The most potent agency to induce the change is exposure of the in-

fusoria to the serum with antibodies of the proper type. The con-

centration of the serum must not, of course, be strong enough to kill

the infusoria. The simplest way to explain these changes is to sup-

pose that the infusoria carry in their cytoplasm plasmagenes of several

serotypes. However, at any one time, one kind of plasmagene out-

numbers the others, and the most frequent kind determines the actual

serotype to which an infusorium belongs. If this plasmagene is sup-

pressed or destroyed by the antibodies in the serum, one of the other

kinds of plasmagenes multiplies instead; the change in the serotype

is the result.

It should be strongly emphasized that the induced transformations

do not resemble the inheritance of acquired characters as imagined by

the early evolutionists (or by Lysenko in Russia). An "acquired char-

acter" is a modification of a part of the body, such as the darkening of

the skin under the influence of sunlight. To be inherited the modified

body part must produce some substance that would reach the genes

in the sex cells, and cause a specific transformation of those, and only

those, genes which determine the traits of the same body part in the

offspring. There is not a trace of evidence that such transformations

occur. Moreover, this view tacitly assumes that there is a one-to-one

correspondence between a gene and an organ or a part of the adult

body; this, however, is not true. Genes determine developmental pat-

terns of the whole organism in different environments. In other words,

90 Heredity, Environment, and Mutation

genes determine which traits will appear in a given environment, but

the genes are not determined by the traits which happen to be realized.

Suggestions for Further Reading

Burks, B. S. 1942. A study of identical twins reared apart under different types

of family relationships. In McNemar and Merrill's Studies of Personality.

McGraw-Hill, New York.

Cattell, R. B. 1950. Personality. McGraw-Hill, New York.

Haldane, J. B. S. 1938. Heredity and Politics. Norton, New York.

Hogben, L. 1941. Nature and Nurture. Norton, New York.

Osborn, F. 1951. Preface to Eugenics. Harper, New York.

Penrose, L. S. 1949. The Biology of Mental Defect. Sidgwich fie Jackson, Lon-

don.

Scheinfeld, A. 1950. The New You and Heredity. Lippincott, Philadelphia.

Stern, C. 1949. Principles of Human Genetics. W. H. Freeman, San Francisco.

The above books and articles are representative of modern points of view con-

cerning the relationships of heredity and environment, particularly in man.

Zirkle, C. 1946. The early history of the idea of the inheritance of acquired

characters and of pangenesis. Transactions of the American Philosophical So-

ciety, Volume 35, pp. 91-151.

This is an excellent outline and analysis of one of the mistaken ideas which

nevertheless played an important role in the history of biology.

Zirkle, C. 1949. Death of a Science in Russia. University of Pennsylvania Press,

Philadelphia.

A documented review of the so-called Lysenko controversy and its background.

Dobzhansky, Th. 1951. Genetics and the Origin of Species. 3rd Edition. Co-

lumbia University Press, New York.

Chapter II of this book deals with the problem of mutation in its relation to

evolution.

Genes and mutations. 1951. Cold Spring Harbor Symposium on Quantitative

Biology, Volume 16.

This volume contains articles by fifty-four authors, representing various ap-

proaches and points of view concerning the study of genes and mutations.

5

Elementary Evolutionary Changes

or Microevolution

Adaptedness and Adaptation. Darwin wrote in 1859: "We see

beautiful adaptations everywhere and in every part of the organic

world." Indeed, the capacity of life to master even most inhospitable

environments is remarkable. During summer months, margins of per-

manent snow fields in high mountains may acquire a pinkish color,

owing to the presence of the alga Sphaerella nivalis. This alga is

adapted to live and reproduce at temperatures close to freezing. In

contrast to it, some algae inhabit hot springs of Yellowstone Park with

temperatures up to 85°C (185°F), which are well above the limit of

toleration for most organisms. The emperor penguins inhabit the

Antarctic ice, and breed during the long and bitterly cold winters.

No nests are constructed; the birds incubate the eggs by placing them

on their feet and covering them by a skin fold which they have on the

belly. Both sexes and all members of a colony compete for the privi-

lege of incubating the eggs, regardless of whether they are the actual

parents.

The universal adaptedness of life to its environment is, next to the

nature of life itself, the greatest problem which biology has to face.

Darwin has attempted to explain the process of adaptation as an out-

come of natural selection impinging upon organic variation. In our

present terminology this means that the conservatism of heredity is

counterbalanced by the dynamism of mutation and sexual reproduc-

tion. Some of the genotypes generated by these processes are adap-

tive in certain environments, and are perpetuated by natural selection;

other genotypes fail to be perpetuated. Some simplest examples of

evolutionary adaptation are discussed in the following pages.

91

92 Microevolution

Resistance of Bacteria to Bacteriophages. D'Herelle (1917) dis-

covered organisms, too small to be seen in ordinary microscopes, which

were called bacteriophages, or simply phages. Under electron micro-

scopes phages appear as tiny spheres, with or without a minute "tail"

(Figure 5.1). If a suspension of bacteriophages is added to a culture

of bacteria, the phages penetrate and multiply inside the bacterial

cells. The bacteria break down, and large numbers of phages are

released, ready to attack new bacteria. Phages, then, are parasites

preying on bacteria. Many studies have been carried out on phages

which attack colon bacteria, Escherichia coli, which occur in the in-

testinal tract of man and of other animals.

If a culture contains many bacteria, one or several bacterial cells

may prove resistant to bacteriophage attack. If the bacteria are kept

on a solid nutrient medium in a Petri plate, the resistant cells multiply

and form colonies of bacteria, which are seen with the naked eye as

white specks on the surface of the medium (Figure 5.2). From such

colonies, new cultures of bacteria may be started which are com-

pletely resistant to the phage. The resistance is a genotypic trait of

the bacterial strain which is retained indefinitely when the strain is

transferred to fresh cultures.

The origin of bacterial strains resistant to bacteriophage attacks has

been studied by many investigators, among whom Burnet (1929),

Luria and Delbruck (1943), and Demerec and Fano (1945) are the

pioneers. Two working hypotheses can be considered. (1) The re-

sistance arises by spontaneous mutation in any bacterial culture, re-

gardless of whether that culture is or is not exposed to the phage.

Suppose that a resistant cell appears about once in ten million to one

hundred million bacterial cell divisions. Cultures which contain hun-

dreds of millions of bacteria are, therefore, likely to have one or several

resistant cells. In the absence of phages, the resistant cells are simply

lost among great masses of normal cells. When a phage is introduced

all normal cells fall prey to its attacks; only the few resistant cells sur-

vive and reproduce, forming the resistant colonies which are easily

picked out. (2) The resistance may, conceivably, be induced in a

small proportion of the cells by their contact with bacteriophages.

The cells that become resistant survive, and the susceptible ones are

destroyed.

According to the first hypothesis, the origin of resistant mutants is

independent of the phages, which act simply as selecting agents, a

sort of sieve which retains only the resistant cells. According to the

second hypothesis, exposure to the phage is the cause which trans-

Ability of Bacteriophages to Attack Bacteria 95

application of the phage will yield a single resistant cell. Statistically,

the result will be that the variance of the number of resistant bacteria

will be greater than the mean number of such bacteria in different

cultures. Experiments have shown that the variances are indeed much

greater than the means. This is a rigorous proof of the validity of the

first hypothesis, that of spontaneous mutation.

Ability of Bacteriophages to Attack Bacteria. If a phage suspen-

sion is applied to a culture of colon bacteria, a mutant line of the

bacteria can be obtained which is resistant to the phage. Several

strains of phages, however, have been discovered which differ in their

appearance under electron microscopes and in certain physiological

characteristics. Now, a line of bacteria resistant to one phage strain

may be fully susceptible to other phages. The resistance in the bac-

teria is specific to the phage strain which was applied as the selector

of the resistant mutant cells.

A line of bacteria which had become resistant to one phage strain

may be made resistant also to other phages. To do so, the bacteria

are exposed successively to different phage strains, and the resistant

mutants are isolated. Bacteria have been obtained which are re-

sistant to several phages. The simplest explanation of this situation is

that the colon bacteria carry many genes, and that the resistance to

different phage strains is due to mutation of different genes. Multiple

resistance is due to occurrence of several mutations.

The selection of resistant mutants adapts bacteria to live in environ-

ments in which phages also occur. The phages meet this situation by

a similar adaptive process. If a large number of phage particles are

added to a bacterial culture which is resistant to that particular phage,

a new strain of the phage may be obtained which can prey on these

bacteria. The new phage strain can be maintained on cultures of

sensitive bacteria. The bacteria may, however, become resistant to

the "new phage" by another mutation, and the phage may overcome

that second resistance by still another mutational step.

The changes in the bacteriophages, like those in the bacteria, seem

explicable on either of the two hypotheses outlined above for bacteria

(page 94). First, mutations which enable phages to attack previously

resistant bacteria may arise spontaneously in phage cultures, and may

be selected when resistant bacteria are exposed to the mixture of mu-

tant and original phages. Second, the ability of the phage to attack

resistant bacteria may be acquired only through contact with such

bacteria. In 1945 Luria demonstrated that the first hypothesis is cor-

rect. Hershey then estimated that the mutations which overcome

96 Microevolution

the resistance of the bacteria occur, on the average, in two to three out

of every billion phage particles.

Resistance of Bacteria to Antibiotics. Discovery of chemothera-

peutic and antibiotic drugs has been a great advance in modern medi-

cine. These drugs kill certain disease-creating microorganisms, often

when applied in extremely small doses. Their efficacy, however, may

be severely reduced by the development of strains of microorganisms

resistant to the antibiotics. Resistant bacteria appear in laboratory

cultures in which the bacteria are kept on nutrient media to which a

certain antibiotic has been added. They can be isolated also from

experimental animals or from human patients who were repeatedly

treated with a chemotherapeutic or antibiotic drug.

A concentration of the antibiotic streptomycin as low as 25 milli-

grams per liter of the nutrient medium stops the growth of the colon

bacteria, Escherichia coli. However, if several billion bacteria are

placed on streptomycin-containing medium, one or several cells con-

tinue to grow and divide, and form colonies from which strains per-

manently resistant to even very high doses of streptomycin may be

established. The situation is quite parallel to that described above for

the development of bacteriophage-resistant strains of the same bac-

teria. Demerec found that mutations to streptomycin resistance arise

in about one out of a billion divisions of the bacterial cells. The mu-

tations are not induced by streptomycin. The role of the strepto-

mycin is that of a selecting agent; it removes vast numbers of non-

resistant cells, and permits only mutant cells to survive.

"Training" of Bacteria. The development of resistance to penicillin

in the bacterium Staphylococcus aureus, studied by Demerec, appears

at first sight less easy to account for on the basis of the mutation-selec-

tion hypothesis. If about one hundred million cells of these bacteria

are placed on a nutrient medium containing 0.1 of an "Oxford unit" of

penicillin per cubic centimeter (milliliter) of the medium, usually

less than ten cells survive and reproduce. But the progeny of these

cells survives well on this medium, and a few cells per billion survive

at concentrations of penicillin up to 0.2 of an Oxford unit per millililer

of the medium. These relatively resistant survivors can now be ex-

posed to still higher concentrations of penicillin. By stepping up the

concentrations of penicillin five times, bacteria are obtained which can

grow in the presence of as much as 250 units of penicillin per milli-

liter of the medium.

This gradual "training" of the bacteria to withstand the presence of

penicillin comes through summation of several mutational steps, each

Resistant Bacteria in Animal Hosts 97

of which taken separately confers only partial resistance on the bac-

teria. Demerec estimates that these mutations occur spontaneously at

a rate of about one per one hundred million bacterial cells (10~8).

Suppose, then, that four or five such mutations must occur to make

the bacteria completely resistant to penicillin. Simultaneous mutation

of two such genes in the same cell is likely to occur about once in a

number of cell divisions expressed by a figure with sixteen zeros

(10~10). Simultaneous mutation of five genes has an utterly negli-

gible probability of occurrence. And yet rigorous selection accumu-

lates such mutations so efficiently that the experiments on induced

resistance are easily reproducible.

Resistant Bacteria in Animal Hosts. Hundreds of reports have been

published in recent years on patients developing drug-fast infections

as a result of treatment with antibiotic and chemotherapeutic drugs.

It is now recognized that most infections treated with streptomycin

must be brought under control by sufficient doses of the drug within a

few days from the beginning of the treatment. If the doses of the

drug used are too small to accomplish this end, the infection is likely

to become streptomycin-resistant; and further treatment with this

antibiotic is ineffective.

When many people are treated with chemotherapy, resistant mu-

tants of pathogenic bacteria acquire so great an advantage over suscep-

tible bacteria that the drugs become effective in fewer and fewer

patients. This is what happened as a result of mass application of

sulfonamide drugs for treatment of certain venereal diseases. The

proportion of cases of disease which failed to respond to sulfonamide

treatment has increased. An unpremeditated experiment of this kind

was also made in certain training centers of the Navy during the

last war. An attempt was made to reduce the incidence of diseases

produced by Streptococcus bacteria by giving to large numbers of

men small doses of a sulfonamid drug. The result was what any

competent geneticist could have predicted: a sharp increase of infec-

tions with sulfonamid-resistant strains of streptococci.

There has been much discussion as to whether the drug-resistant

bacteria arise in the treated patients, or are always present in bacterial

populations. This is a spurious problem. Any mutation has a certain

probability of occurrence; the application of the antibiotics does not

influence the rate of origin of the resistant mutants. Nevertheless, the

antibiotics are responsible for the spread of such mutants, giving them

an adaptive advantage which they do not possess outside the environ-

ments in which the drugs are present.

98 Microevolution

Reversibility of the Effects of Mutation and Selection. Mutant bac-

teria resistant to bacteriophages or to antibiotics are superior to sensi-

tive bacteria in environments in which bacteriophages or antibiotics

are present. Sensitive bacteria are killed off, the resistant ones sur-

vive. Yet resistant mutants are not induced by the phages or the anti-

biotics; they arise regardless of whether the environmental selecting

agents favor them or not. Why, then, are most colon bacteria found

outside of laboratories still susceptible to bacteriophage attacks and

sensitive to streptomycin? Why have the resistant mutants not

crowded out the sensitive genotypes? The theory leads us to infer

that the resistant mutants must in some respects be at a disadvantage

compared to sensitive bacteria in the absence of phages and anti-

biotics.

This theoretical inference is strikingly verified in some experiments.

Close to 60 per cent of the streptomycin-resistant mutants in colon

bacteria are also streptomycin dependent; these mutants are unable

to grow on culture media free of streptomycin. A substance which is

poisonous to normal sensitive bacteria is essential for life of the re-

sistant mutants! E. H. Anderson has shown that some bacteriophage-

resistant strains of colon bacteria require for growth certain food sub-

stances which are not needed for the growth of sensitive bacteria. The

resistant mutants will be wiped out in environments in which the

required foods are not available.

Just what puts some mutants at a disadvantage in competition with

"normal" types is unknown at present. Some streptomycin-resistant

mutants can grow without streptomycin, and some phage-resistant

mutants are not known to possess special food requirements. It should

be noted, however, that advantages and disadvantages which are

easiest to discover are of the all-or-nothing type: all streptomycin-

sensitive bacteria are killed in the presence of streptomycin, and all

streptomycin-dependent bacteria die in the absence of this substance.

But the advantages and the disadvantages may also be quite small

and yet very effective in the long run. Suppose that a mutant grows

slightly slower than the parental type in a certain environment. The

mutant will then be quite rare, and yet it may not be easy to discover

the nature of the handicap which keeps it rare.

Where the advantages and disadvantages are of the all-or-none type,

the importance of the environment in selecting the best-adapted geno-

types can be demonstrated with remarkable clarity. Demerec and his

collaborators showed that the changes from streptomycin sensitivity

to resistance and dependence are reversible. A large number of colon

Genetics and Epidemics 99

bacteria of a streptomycin-dependent strain are placed on a Petri plate

with nutrient medium free of streptomycin. Only the mutant cells

which can grow without streptomycin survive and form colonies of

streptomycin-independent bacteria. The mutation from streptomycin

dependence to independence occurs about forty times per one billion

divisions of the bacterial cells.

Virulence of Pathogenic Microorganisms. Disease-producing mi-

croorganisms often lose their ability to infect their normal hosts after

prolonged cultivation on artificial media in laboratories, but the viru-

lence of such "degenerate" bacteria may sometimes be restored. For

this purpose a large number of the bacteria which lost their virulence

are inoculated in a host susceptible to the disease. If the bacteria

are able to multiply in such a host, a new virulent strain may be

isolated.

Fowl typhoid, a disease often fatal to chickens, is caused by the

bacterium Salmonella gallinarum. In an experiment by Gowen, an

originally highly virulent strain of these bacteria lost its virulence after

some months of cultivation in the laboratory. Inoculation of about

100,000 bacteria of the virulent strain in a bird results in disease which

is fatal to some 70 per cent of the birds; the same dose of avirulent

bacteria produces no disease. A thousand times greater (100,000,000)

dose of avirulent bacteria results in the death of only 34 per cent of

the chickens. The full virulence, however, is restored by inoculating

about two billion avirulent bacteria into each of several chickens;

after a week, new cultures of the bacteria are re-isolated from the in-

fected birds. Such "passage" of the bacterial strain through living

hosts was repeated six times. New virulent strains of bacteria were

obtained.

The simplest interpretation of the loss and reacquisition of the

virulence is that different genotypes are advantageous in bacteria

which live as parasites in living hosts, and multiply in laboratory test

tubes. When bacteria grow on laboratory media, mutants are selected

which enable them to live most successfully in this artificial environ-

ment, even though the mutants lose the ability to invade a host's body.

Conversely, when laboratory strains of bacteria are inoculated in a

host's body, the mutants which restore the ability to overcome the re-

sistance of the host acquire a clear survival advantage.

Genetics and Epidemics. Mankind has always dreaded the recur-

rent scourge of epidemics. Some 25 million people perished from

black death in Europe in 1347 and 1348, between a quarter and a

100 Microevolution

third of the population. Outbreaks of infections occur also in wild

and in domestic animals and plants.

We have seen that mutations in pathogenic microorganisms make

them more or less virulent, and more or less sensitive to antibiotics.

There exist also genetic variations in the susceptibility to diseases

among the hosts which the pathogens attack. Perhaps the clearest

example is the resistance of varieties of the common onions to attacks

by the smudge fungus, Colletotrichum. Onion varieties which have

white bulbs are easily infected with the fungus; cream-colored or

yellow bulbs are slightly susceptible; and deeply colored red or purple

bulbs are resistant. The resistance is due to the presence in the scales

of the colored bulbs of catechol and protocatechuic acid, substances

which can be shown to be poisonous to the spores of the fungus.

Walker and his students have shown that the color of the bulbs, the

presence in them of catechol and protocatechuic acid (which are

colorless substances), and the resistance or susceptibility are manifes-

tations of the same genes. The gene 7, when homozygous (77), makes

the bulbs white and susceptible to the infection. The heterozygotes,

li, are yellow and only slightly susceptible, whereas the homozygotes,

it, are deeply colored and resistant.

There is no doubt that genetic variations in the susceptibility to

infection occur also among higher animals and among men. Gowen

has described strains of chickens some of which were more resistant

than others to the fowl typhoid, resulting from infections with Salmo-

nella (see above). Genetic resistance interacts, of course, with the

immunity to certain diseases acquired by individuals as a result of

the formation of protective antibodies in their blood. The formation

of protective antibodies may be induced by a previous infection or,

artificially, by vaccination. The genotype determines, as we know,

the responses of the organism to the environment. The actual resist-

ance or susceptibility of an individual (his phenotype) will depend

upon his genetic constitution as well as upon his previous contacts with

the disease.

An epidemic, then, will affect at first those individuals in the host's

population which are phenotypically and genetically most susceptible

to the infection. An epidemic may start when a strongly virulent

strain of the pathogen appears. The spread of the infection will be

more and more rapid as the number of its victims increases, but the

point will be reached when a majority of the highly susceptible indi-

viduals will be either dead or recovered from the disease. An epi-

demic grows so long as each victim transmits the infection, on the

Rust Fungi Attacking Wheat 101

average, to more than one new victim. When the frequency of the

transmission drops below one, the epidemic subsides. The outcome

of an epidemic may be a genetic change in the composition of the

populations of both the host and the parasite.

Rust Fungi Attacking Wheat. As far as man is concerned, the fore-

going picture of the genetic background of epidemics is rather hypo-

thetical. It is, however, borne out by observations particularly on the

diseases of certain crop plants. The rust fungus, Puccinia graminis

tritici, is responsible for one of the most serious diseases which affect

wheat plantings. This species of rust contains some 200 varieties

(called "physiological races" by the students of these fungi). The

rust varieties are distinguished from each other by their ability to in-

fect different varieties of wheat. Most wheat varieties are susceptible

to infection by some, but resistant to other varieties of the rust.

The spores of the rust are distributed by winds and air currents.

Since the degree of infestation of wheat fields by the rusts influences

the yield of the crop, annual census of the relative frequencies of the

spores of different rust varieties is made in the United States and in

other wheat-growing countries. The incidence of the different rust

varieties undergoes striking changes with time (Table 5.1).

TABLE 5.1

FREQUENCIES (IN PER CENTS OF THE TOTAL) OF THE SPORES OF THE DIFFERENT

VARIETIES OF THE RUST FUNGUS Puccinia graminis tritici IN THE UNITED STATES

(After Stakman and Others.)

Variety Number

Year

1930

1931

1932

1933

1934

1935

1936

1937

1938

1939

1940

1941

1942

1943

1944

11

17

21

34

36

38

49

56

4

0.3

7

0.6

36

SO

20

0.2

22

0.6

4

2

28

IS

25

1

5

1

2

0.9

102 Microevolution

The changes in the relative frequencies of the rust varieties are

largely due to the introduction of new varieties of wheat for mass

planting by the farmers. Wheat breeders are constantly trying to

create new kinds of wheats resistant to infection by the rusts. They

are often successful, to the extent that new wheat genotypes are ob-

tained by hybridization and selection which are resistant to the rust

varieties that are prevalent at the time when the breeding work is

done. These new kinds of wheat are then multiplied and planted by

many farmers. At first, the yields of wheat are satisfactory. But the

new wheats begin to be attacked by rust varieties which were hitherto

unknown or rare, and soon the rust infestations become so heavy that

the wheat yields decline. Eventually it becomes necessary to use

another assortment of wheats for planting.

For example, the outbreak of rust variety 56 in 1934 and thereafter

(see Table 5.1) seems to be connected with large-scale planting in the

United States of the wheat variety called Ceres. In 1925 this high-

yielding wheat was planted on a large scale. About ten years later

Ceres was so heavily attacked by rust that its plantings rapidly shrank

in number. This is a genetic and epidemiological experiment con-

ducted on a grand scale. Introduction by man of a new host plant

initiates a process of selection of genotypes in the parasite which can

invade this host, with the result that the adaptedness of the host geno-

types declines. An unstable equilibrium is temporarily restored by a

genotypic change in the host.

Insect Pests Resistant to Poisons. Experiments on adaptation

through selection of useful mutants have been made chiefly with bac-

teria and other microorganisms because the experimenter can easily

manage huge numbers of individuals and of generations, which would

be out of the question with larger creatures. Similar adaptive processes

occur also in higher organisms, although usually more slowly. Many

insecticides, poisons which kill insect pests, have been discovered. A

particularly powerful insecticide is dichloro-diphenyl-trichlorethane,

popularly known as DDT. This substance is highly poisonous to most

insects in such low concentrations as to deserve being placed among

the "wonder drugs." The struggle of man against insect pests has

nevertheless not ended, owing to the emergence of DDT-resistant

varieties of insects.

The housefly, Musca domestica, is often controlled by DDT sprays.

Yet Missiroli and Weismann reported independently in 1947 that

houseflies from certain localities in Italy and in Sweden, respectively,

had become relatively resistant to DDT. The concentrations of DDT

Insect Pests Resistant to Poisons 103

which in previous years were sufficient to destroy most of the house-

flies in these localities no longer were effective when applied in the

customary way. Since 1947 similar reports have come from Denmark,

England, Greece, Egypt, Peru, Venezuela, Mexico, and parts of the

United States as widely separated as New Jersey, Florida, California,

and Oregon. In many places DDT can no longer be relied upon to

bring housefly infestations under control.

Strains of houseflies resistant to DDT have been obtained also in

laboratory experiments. Lindquist and Wilson exposed a normal sen-

sitive strain to a dose of DDT which killed about 90 per cent of the

individuals, and used the survivors as parents of the next generation.

A significant resistance to DDT appeared after three generations of

such selection.

There is no agreement among different investigators about how

permanent is the resistance once acquired by a fly strain. In some

instances the resistance has been retained for ten or more genera-

tions without renewed exposures to DDT; in others the resistance has

been alleged to become weaker or lost. This situation may be inter-

preted in several ways. It is possible that the DDT resistance is due

to a single mutation, analogous to that which confers the streptomycin

resistance on bacteria (page 98). It may also be that the resistance

is produced by a combination of several genes which are normally

present in most fly populations. In either case, a "resistant" strain

may actually be a mixture of more or less resistant and of sensitive

genotypes. The average resistance may vary greatly from generation

to generation, and it may be more or less fixed or changeable in dif-

ferent strains.

Theoretically, we might expect the resistant houseflies to be adap-

tively inferior to the sensitive ones in the absence of DDT treatments.

Indeed, some investigators have found that resistant flies take more

time to develop than the normal ones. If confirmed, this difference

may be very significant, since a rapid development may be advan-

tageous to the flies in their normal habitats. Other investigators have

found indications of important physiological differences between sen-

sitive and resistant flies.

The emergence of strains of insect pests resistant to various insecti-

cides has assumed, in recent years, the proportions of a major problem

to those concerned with pest control. Resistances to DDT treatments

have been described in several species of mosquitoes, in filter flies

(Psychoda), cockroaches, body lice, bedbugs, and bark beetles

(Scolytus). Even before the advent of the DDT, strains resistant to

Deleterious Character of Most Mutations 105

delian gene, the dark variety being usually dominant to the light one.

The light varieties are the "normal" ones, in the sense that until re-

cently the moths were represented in nature by light individuals, the

dark ones being exceptional or altogether unknown. But beginning

in the middle of the nineteenth century, the dark varieties have been

recorded as occurring more and more often, especially in the vicinity

of large industrial centers. In many industrial regions the incidence

of dark varieties has grown rapidly from decade to decade; some of

the populations now consist of dark individuals, with or without an

admixture of light ones.

The most satisfactory interpretation of the spread of the melanic

varieties has been suggested by Ford, who believes that the dark va-

rieties are in general more vigorous than the light ones. In one spe-

cies, Boarmia repandata, he was able to demonstrate the superior vigor

of the melanic forms experimentally. Why, then, are the dark varieties

crowding out the normal light ones only in industrial regions and not

everywhere? Ford's answer is that the light varieties compensate for

the deficient vigor by being protectively colored. In other words, they

are camouflaged by matching the coloration of their surroundings, and

escape being detected and eaten by their enemies, insectivorous birds,

more often than do the more conspicuous dark varieties. The situa-

tion is different in industrial regions, where the countryside is con-

taminated by soot and other waste products. Here, on darker back-

grounds, the light varieties are no longer camouflaged, whereas the

dark varieties may more nearly match the colors of their surround-

ings. Industrialization of a region removes, then, the disadvantages

of the dark varieties, which spread and eventually supplant the light

ones.

Deleterious Character of Most Mutations. Most mutants which

arise in any organism are more or less disadvantageous to their pos-

sessors. The classical mutants obtained in Drosophila usually show

deterioration, breakdown, or disappearance of some organs (Figures

4.4 and 4.5). Mutants are known which diminish the quantity or

destroy the pigment in the eyes, and in the body reduce the wings,

eyes, bristles, legs. Many mutants are, in fact, lethal to their possessors.

Mutants which equal the normal fly in vigor are a minority, and mu-

tants that would make a major improvement of the normal organiza-

tion in the normal environments are unknown.

And, yet, modern theories of evolution consider the process of mu-

tation to be the source of the raw materials from which evolutionary

changes may be constructed. The deleterious character of most mu-

106 Microevolution

tants seems to be a very serious difficulty. A more careful considera-

tion shows, however, that the difficulty is much less formidable than

it may appear at first sight.

Are the mutations which make the colon bacteria resistant to bac-

teriophages or to streptomycin harmful or useful? Is the resistance to

DDT sprays deleterious or advantageous to the houseflies? Obviously,

these questions are meaningless when the nature of the environments

in which the mutants are placed is not specified. On media containing

bacteriophages or streptomycin, the resistance is necessary to the bac-

teria, since the carriers of the non-resistant genotypes are destroyed.

Flies or scales resistant to the DDT or to hydrocyanic gas have evident

advantages in localities sprayed or fumigated with these poisons, but

the same genotypes are disadvantageous when phages, antibiotics, and

insecticides are absent.

Useful Mutations. Useful mutations, therefore, are known. They

are in fact not uncommon when organisms are placed in environ-

ments other than those in which these organisms usually occur, but

in environments in which the species normally lives useful mutations

are not observed. A closer consideration will show that this is, indeed,

as it should be. Every kind of mutation that a biologist observes in

his experiments has a certain probability of occurrence, and it has

occurred repeatedly in the history of the species. The mutations

which improve the adaptedness of the species to its normal environ-

ments were tried out by natural selection, and have become incor-

porated into the "normal," that is, frequently met with, genotypes. The

"normal" genotypes are compounded of just such useful mutations.

Hence the mutations that take place are deleterious.

But when the environment changes, the adaptedness of the organism

is disrupted, and the harmony between the environment and its in-

habitants can be restored only through a genotypic reconstruction.

Some mutations prove useful in new environments and in combina-

tion with new components of the species genotype. The mutant genes

replace the genes that were "normal" in the old environment, and

become the new adaptive "norm." Although the old genes may re-

appear from time to time by mutation, they will now be deleterious

and will be eliminated. This is what happens when colon bacteria

resistant to streptomycin are kept on media with high streptomycin

content. The mutants which restore the sensitivity to strepomycin

are destroyed.

The Price of Adaptation. Many mutants which are deleterious in

a certain environment may be useful in altered circumstances. It does

The Price of Adaptation 107

not follow, however, that every mutant which occurs in any species

must necessarily be useful in some environment; numerous mutants

in Drosophila and in man produce hereditary diseases which can

hardly be useful in any environment. It is difficult to imagine an

environment which would make hemophilia or the amaurotic idiocy

useful.

There is nothing surprising in the appearance of such adaptively

worthless mutants. A mutation is essentially a dislocation taking place

in the delicate self-reproducing mechanism of the gene. Random

changes in any complex mechanism, such as a watch or an automobile,

are more likely to injure than to improve it; and yet watches and auto-

mobiles are changed for the better, step by step, in their historical

development, just as the organisms are in their evolutionary develop-

ment. The improvements in non-living mechanisms are due to the

skill of the engineers; those in living beings are established by natural

selection perpetuating the useful mutants and gene combinations, and

failing to perpetuate the useless ones.

But why, it may be asked, should useless mutations be produced at

all? It would be vastly better if the organisms produced only useful

mutations where and when needed. A little thought will show how

naive such a question really is. The normal colon bacteria produce

bacteriophage-resistant and streptomycin-resistant mutants, a great

majority of which are lost, because they are useless in the absence of

phages and of streptomycin. In order to produce only phage re-

sistance in the cultures to which phages are applied, only streptomy-

cin resistance in the cultures which are brought in contact with strep-

tomycin, and neither mutant in cultures which remain in normal en-

vironments, bacteria would have to possess a prescience of the future.

When the phages attack it is too late: the non-resistant bacteria perish.

All kinds of mutants are produced regardless of whether any of them

will prove useful. A living species that would suppress the mutation

process, by making the gene reproduction absolutely accurate, might

gain a temporary advantage. In a static environment such a non-

mutable species would not beget the useless and harmful mutants,

but when the environment changed and made some of the mutants

better fit than the old norm to survive, a non-mutable species would

be the loser. In the long run, mutability is a useful trait, even though

most mutants are harmful in any one environment. Herein lay the

fundamental error of the belief in the inheritance of acquired traits:

it assumed that the genes are made to mutate in a determined direc-

tion by every change in the development of the body.

108 Microevolution

Suggestions for Further Reading

Darwin, Ch. 1859. On the Origin of Species by Means of Natural Selection.

This is one of the greatest books of all time, and an acquaintance with it is a

"must" for every student of evolution. There are numerous editions of this work.

The first edition (1859) has recently been reprinted by the Philosophical Library,

New York, with a foreword by C. D. Darlington (1951).

Catcheside, D. G. 1951. The Genetics of Microorganisms. Pitman, New York.

Braun, W. 1953. Bacterial Genetics. Saunders, Philadelphia.

The books of Catcheside and of Braun review the experiments on the genetics

of microorganisms.

Dunn, L. C. (Editor). 1951. Genetics in the Twentieth Century. Macmillan,

New York.

This is a symposium held in 1950 to commemorate the fiftieth anniversary of

the rediscovery of Mendel's work and of the birth of genetics. Several papers in

this symposium are pertinent to the topics discussed in the present chapter, espe-

cially Genetic studies in bacteria, by J. Lcderberg; Genetics and disease resistance,

by J. W. Gowen; Genetics and plant pathology, by J. C. Walker.

6

Natural Selection

and Adaptation

It is a commonplace observation that every living being is so con-

structed that it is able to live in a certain environment (Figure 6.1).

A fish is adapted to live in water, a bird is an efficient flying machine,

a cow and a deer have digestive organs which enable them to feed on

herbage and foliage, the human mind permits man to acquire and

transmit culture. The origin of this apparent purposefulness of bio-

logical organization is a riddle which several generations of biologists

have attempted to solve. Some have taken the easy way out by sup-

posing that every living species is endowed by the Creator with those

features which it needs in order to live in the habitats in which it is

actually found. This is, however, a spurious solution; it implicitly

blames the Creator also for all the .imperfections and all the sufferings

found in human and biological nature.

So far the only scientifically tenable solution of the riddle was pro-

posed by Darwin in 1859. According to Darwin, organisms become

adapted to the environment in the process of evolution; this process is

controlled by natural selection of genetic variants which are relatively

better fitted than others to survive and to reproduce in certain environ-

ments. The theory accepted by a majority of modern evolutionists is

clearly derived from Darwin's theory; however, it is just as clearly not

identical with the Darwinian prototype. The changes in the theory

are due to the mass of new evidence discovered by biologists since

1859. The modern theories are often referred to as neo-Darwinism,

which would be a good name if it were not for the fact that it was

applied also to the theories developed by Weismann and others around

1900, and they are as different from the modern ones as they were

from Darwin's. Synthetic theory and biological theory of evolution

109

History of the Idea of Natural Selection 111

are better names, because they emphasize that modern theories are a

result of synthesis of the findings of many biological disciplines-

genetics, systematics, comparative morphology, paleontology, embry-

ology, ecology, etc.

History of the Idea of Natural Selection. Some historians believe

that a germ of the theory of natural selection is as old as the myths

of the Greek philosopher Empedocles (fifth century B.C.) and the

Roman poet Lucretius (first century B.C. ). According to these myths,

parts of human bodies, heads, trunks, arms, and legs appeared in the

world before complete human beings. These isolated parts proceeded

to combine at random, giving rise to many inviable combinations which

were lost, and to complete human bodies which survived and repro-

duced their likes.

According to Darwin himself, the idea of natural selection was

suggested to him by the Essay on the Principles of Population, pub-

lished in 1798 by the English country parson and amateur sociologist

and mathematician T. R. Malthus. Malthus pointed out that human

populations tend to increase in numbers in geometric progression; that

growing populations sooner or later outrun their food supply; that

more children are born than live to maturity, because of the "struggle

for existence"; hunger, war, and disease keep the population numbers

in check. Darwin saw that this is obviously true for any living species,

from a microbe to the elephant. Indeed, an oyster produces up to

114,000,000 eggs in a single spawning; under favorable conditions,

some bacteria can double in number in less than half an hour. Higher

animals, mammals and birds, produce only few young, but on the aver-

age always more than two per pair of parents. Given enough time,

the number of individuals of any species could become too large for

the earth to support them.

In reality, the populations of most species oscillate within rather

narrow limits, and only rarely do some populations grow as fast as

their reproductive powers might permit. The struggle for existence

causes a part, and often a very large part, of the progeny to die before

reaching sexual maturity. Living beings "fight" against the physical

environment—excessive cold, heat, dryness, wind, etc. They are de-

stroyed by predators, parasites, and diseases. And they "compete"

with individuals of their own and other species for food, for shelter,

and for mates.

Darwin saw further the relation between the Malthusian struggle

for existence and the fact that agriculturists have since time immemo-

rial improved the quality of domestic animals and plants by selecting

112 Natural Selection and Adaptation

for reproduction those individuals which possessed the qualities which

were useful or desirable to the breeders. Now, the survivors in the

struggle for existence are likely to differ in their genotype from the

victims which succumb in this struggle. Provided that the survivors

differ in heredity from those who do not survive, the struggle for

existence will automatically result in natural selection of the fitter

genotypes. The hereditary endowment of the succeeding generations

will differ from that of the preceding generations in the direction of

superior fitness.

Competition and Cooperation. The logic of Darwin's reasoning was

unassailable. Nevertheless, he and his theories were attacked by those

who believed that the recognition of man as a part of nature is sub-

versive to religion and insulting to human dignity. In the ensuing

polemics some unfortunate overstatements were made. Herbert

Spencer said that the struggle for existence led to the "survival of the

fittest." The use of the superlative gave a subtle overemphasis to the

supposed struggle and competition—as though only one fittest survived

and all the rest died. Spencer's statement sounded very much like

the concept of Superman, advanced by the German philosopher

Nietzsche, which in turn played a role in the development of racist

and Nazi ideas.

Phrases such as "struggle for existence," "nature red in tooth and

claw," "eat or be eaten" were very freely used, especially by popular

writers on evolution, in the late nineteenth and early twentieth cen-

turies. This phraseology seemed to appeal to the emotions of many

people of those days. With the development of genetics and with a

change of the intellectual climate during the current century, it began

to be realized that the fierceness of the struggle for existence leading

to natural selection was greatly exaggerated. It is simply the fit, rather

than the "fittest," who survive.

Evolutionary success is determined by the ability of the carriers of

a given genotype to transmit their genes to the greatest possible num-

ber of individuals in the following generations. "Struggle," "competi-

tion," and like expressions have a metaphorical meaning when applied

to the process of natural selection which should not be confused with

their usage when applied to human affairs. Thus trees "struggle"

against the danger of being felled by wind by developing stronger root

systems; mammals and birds "struggle" against cold by developing

heat insulation, temperature regulation, or by remaining dormant dur-

ing winter months; desert plants "struggle" against dryness by having

leaves transformed into spines. Plants and animals "compete" for food

Johannsen and Pure Lines 113

when food is scarce, but they do not necessarily fight against one an-

other.

Among the early critics of Darwin, the English writer, Samuel

Butler, and particularly the Russian political philosopher, Kropotkin,

sought to replace Darwin's theory of "struggle" by one which made

cooperation and mutual help the principal agents of evolution. Re-

cently this idea was developed further and modernized by Allee and

by Ashley Montagu. As a corrective against the abuse of Darwinism

by political propagandists, this emphasis on the importance of co-

operativeness in natural selection is very useful. Indeed, pugnacity

and aggressiveness are often less conducive to biological success than

is inclination to "live and let live" and to cooperate with other indi-

viduals of the same and of other species. The fact is that both com-

petition and cooperation are observed in nature. Natural selection is

neither egotistic nor altruistic. It is, rather, opportunistic: life is pro-

moted now by struggle and now by mutual help. Thus parental care

and defense of the offspring are obviously advantageous for the sur-

vival of the species; and yet members of an established colony of a

certain species of ant in Australia destroy the daughter colonies as

soon as the latter appear on the surface of the ground in the vicinity

of the parental nest.

Johannsen and Pure Lines. Less spectacular but more important

than the misunderstanding of the relative importance of competition

and cooperation in natural selection was another difficulty of Darwin's

theory. As pointed out above, Darwin realized quite clearly that

natural selection will produce permanent improvements only provided

that the surviving "fit" differ in genotype from the eliminated "unfit."

If the individuals which breed and those which are prevented from

breeding are alike in genetic endowment, their progenies should be

alike and selection should be ineffective. This matter was clarified

long after Darwin's death by the Danish botanist Johannsen (1857-

1927).

Johannsen (1909) started his experiments with garden beans by

choosing seeds of different sizes and weights from a commercial va-

riety, and keeping the progenies of the plants grown from each seed

separately (Figure 6.2). The garden bean is a plant species which,

like Mendel's pea, reproduces mostly by self-pollination. Progenies

obtained from a single individual of such species by self-fertilization

are called pure lines, and individuals which belong to the same pure

line usually have similar genotypes. The pure lines obtained by

Johannsen from larger beans produced, on the average, larger seeds

Gene Frequency and Gene Pool 115

variations in the environmental conditions under which the plants

grow; they are within the norm of reaction of the genotype of the

pure lines.

Johannsen's work was confirmed by other investigators working

with pure lines and with clones. A clone is the offspring of a single

individual obtained by asexual means, such as simple fission and bud-

ding. Clones can be easily obtained in most microorganisms and in

some plants and animals which can reproduce asexually. Selection is

ineffective in a clone (unless mutation intervenes; see below).

In effect, the work of Johannsen proved only that acquired char-

acters are not inherited (see Chapter 4). Indeed, phenotypic differ-

ences between genotypically uniform individuals are acquired through

environmental modification, but in the 1910's these results were inter-

preted as a blow to Darwin's theory of natural selection. For Johann-

sen found that selection was effective only in the original material,

which was a mixture of several genotypically different pure lines.

Selection, then, has done nothing "creative"; it has merely isolated

some of the genotypes which were already present in the original mix-

ture. In the absence of pre-existent genotypic differences, selection is

powerless to effect evolutionary changes.

The modern solution of the foregoing difficulty is simple enough.

Selection operates with genetic raw materials supplied by mutation

and sexual reproduction. Mutation and sex produce an abundance of

genetic variability, interacting with which selection becomes a creative

process (see Chapter 14). But before this solution could be reached,

another, and apparently even more formidable, difficulty of the Dar-

winian theory had to be removed.

Gene Frequency and Gene Pool. According to the "blood" theory

of heredity, which was quite generally accepted in Darwin's day, the

heredity of a child is a solution, or an alloy, of the heredities of the

parents. Consider, then, what happens in a population which contains

genotypes that produce, in a certain environment, individuals of a

middle stature, genotypes which produce tall and short people, and

a few genotypes which produce giants and dwarfs. In the genetic

sense, a population, a Mendelian population, is a community of sexu-

ally reproducing individuals among whom marriages are concluded.

Suppose, furthermore, that our Mendelian population is panmictic

with respect to stature; in other words, that the likelihood of an indi-

vidual's marrying a partner of any stature is proportional to the rela-

tive frequencies of these statures in the population. It is easy to see

that, in such a panmictic population, individuals who are taller than

Gene Frequency and Gene Pool 117

the average will often marry partners who are shorter than themselves,

whereas persons who are shorter than the average will usually marry

mates who are taller than they are.

If the blood theory of heredity were true, the heredity of children

would be a compromise between the heredities of their parents. A

panmictic population would then become less and less variable in

stattrre, and in other characters, as one generation followed the other.

In other words, a sexual panmictic population would become more

and more uniform, and eventually it could reach the point where all

individuals composing the population would be uniform in hereditary

endowment. Such a genetically homogeneous population could be

called a pure race.

Darwin realized that a necessary consequence of the acceptance of

the blood theory of heredity was the belief that sexual reproduction )

tended to level off any genetic variability that might be present in a /r

sexual population. And yet, if evolution is to occur, populations musr ('

always possess a supply of hereditary variability. Darwin admitted i

that the contradiction was insoluble in his time. At present it is V

known that the contradiction was spurious, because, as shown so

clearly by Mendel, heredity is transmitted by genes rather than by

"blood." The problem was solved in 1908, independently by Hardy in ty

England and by Weinberg in Germany.

Suppose that a sexually reproducing population lives isolated from *3""

other populations of the same species, for example, on an oceanic

island. Suppose, furthermore, that the population consists of indi-

viduals homozygous for a gene AA, and of individuals homozygous

for aa. Let the proportion of AA individuals in the population be

represented by a fraction q, and the proportion of aa by (1 — q). If

the population is panmictic with respect to this gene, there will be q2

matings AA X AA, which will produce A A children; (1 — q)2 matings

aa X aa, which will give aa children; and 2q(1 — q) matings AA X oo,

which will give heterozygous, Aa, progeny.

Instead of considering the genotypes of the parents who mate, it is

more convenient to consider the sex cells which they produce. Sup-

pose, then, that all individual members of a sexual population are

equally fertile and contribute equal numbers of sex cells or gametes

to the "gene poof of the population. If the population is panmictic,

these sex cells unite at random, so that A and a eggs are fertilized by

A and a sperms in proportion to their frequency in the gene pool. Let

the fraction of the sex cells which carry A be q, and the fraction of

118 Natural Selection and Adaptation

sex cells a in the gene pool be (I — q). The progeny, consequently,

will be:

Eggs

qA (1 - q)a

e

j. a

* ?•

In other words, there will be respectively q" and (1 — t/)2 of

homozygous AA and aa, and 2
important problem is what the distribution of the genotypes will be

in the next generation. Let us again consider the composition of the

gene pool. The homozygotes AA and aa produce only A and a

gametes, whereas the heterozygotes Aa produce equal numbers of A

and a sex cells. The total frequencies of the A and a gametes in the

gene pool will accordingly be:

92.4.4

9(1 - q}Aa

(1 — q)q Aa

(1 - 9)2aa

a=

- q) + (1 - q)2 = q - q2 + I - 9q + q2 = 1 - q

The frequencies of the genes A and a in the gene pool will thus re-

main constant at the levels q and (1 — q ) , and the distribution of the

genotypes among the zygotes in the population will be:

q2AA:2q(l - q)Aa:(l - q)2aa

This is the Hardy-Weinberg formula, which describes the equilibrium

condition in a sexually reproducing population and is the corner-

stone of modern population genetics. The gene frequencies, q and

(1 — q), remain constant in sexually reproducing populations. Con-

sider again the example of a human population consisting of persons

different in stature. Some people are tall and others short, owing to

the presence in the population of genes affecting the stature. Each

of these genes, which may be called A and a, B and b, etc., has a cer-

tain frequency in the gene pool. In the absence of any of the factors

changing the gene frequencies described below, the population will

always have some tall, average, and short persons, as well as some

giants and dwarfs.

Factors Which Change Gene Frequencies. The Hardy-Weinberg

theorem describes the statics of a Mendelian population. If all gene

Adaptive Value 119

frequencies in all populations remained constant, evolution would not

take place. Evolution may be defined in a most general way as a

change in gene frequencies. In reality, agents are known which may

alter the gene frequencies in populations. The dynamic forces are as

follows:

(1) Mutation. If the allele A changes into a, or vice versa, the fre-

quencies q and (1 — q) may become altered.

(2) Selection. The Hardy-Weinberg theorem assumed that the

carriers of all genotypes contribute equally to the gene pool of the

following generation. This is not necessarily the case. Some geno-

types may be more fecund than others, or they may have different

viabilities, different longevities, some may be sexually more active than

others, etc. Most differences of this kind cause variations in the adap-

tive values of carriers of genotypes, and may bring about changes in

gene frequencies.

(3) Migration. Relatively more carriers of the gene A than of a

may emigrate from or immigrate into a population.

(4) Genetic Drift. The assumption implicit in the Hardy-Weinberg

theorem is that the population consists of so many individuals that

chance variations in gene frequencies are negligible. Strictly speak-

ing, this condition could obtain only in ideal, infinitely large popula-

tions. In reality, all populations are finite, and some populations are

small.

In deriving the Hardy-Weinberg formula it was assumed that the

populations are panmictic. For some natural populations this assump-

tion does not hold true. For example, many higher plants and some

hermaphroditic animals reproduce by self-fertilization. In wheats,

oats, beans, peas, and some other crop plants the ovules are fertilized

mostly by pollen from the same flower. Inbreeding, mating of close

relatives, and self-fertilization lead to increased frequencies of homo-

zygotes, and decreased frequencies of heterozygotes, compared to

panmictic populations. Gene frequencies are not usually altered by

deviations from panmixia.

Adaptive Value, Darwinian Fitness, and Selection Coefficient. The

carriers of some hereditary endowments are more viable and reach

the reproductive age more often than do carriers of other genotypes.

Representatives of different genotypes vary also in fecundity (num-

bers of young born, or of eggs or seeds produced), in sexual activity,

in the duration of the reproductive period, and in other properties

which favor the reproductive success of the organism. The viability ,

and the reproductive success determine the contribution which the /

120 Natural Selection and Adaptation

carriers of a genotype make to the gene pool of the next generation

of the species or of a population. This contribution is a measure of

the adaptive value, or Darwinian fitness, of the genotype. Thus, if a

genotype produces, on the average, 100 surviving offspring where an-

other genotype produces 90 offspring, the adaptive value of the former

is 1.00 and of the latter 0.90. It is also said that the second genotype

is opposed by a selection coefficient of s = 0.10.

An Example of Estimation of the Speed of Selection. Haldane,

Fisher, Wright, and other investigators have analyzed mathematically

the changes which occur in populations under the influence of various

forms of selection. It will be sufficient here to consider a single ex-

ample. About 70 per cent of the American white population feel a

bitter taste in certain solutions of the PTC substance, whereas about

30 per cent are non-tasters. The non-tasters are homozygous for a

recessive gene f; the tasters are either homozygous or heterozygous

for the dominant T (TT or Tt, cf. Chapter 2, page 30). Since people

do not usually know whether they are tasters or non-tasters of PTC,

the population may be assumed to be panmictic with respect to this

trait. According to the Hardy-Weinberg formula, the frequencies of

the tasters and non-tasters should be qzTT : 2q(\ - q)Tt : (1 -

q)2tt. Since the non-tasters, tt, constitute 0.3 (30 per cent) of the

population, the frequency of the gene t in the gene pool should be

(1 — q) - VO-3 = 0.548. The frequency of the gene T in the gene

pool is, then, q = 1 — 0.548 = 0.452. The frequencies of the three

genotypes will, then, be:

q*TT = 0.204; 2?(1 - q)Tt = 0.495; (1 - q)2U = 0.300

Assume now that all non-tasters, tt, are removed from the popula-

tion or are prevented from reproduction. The genotype tt will, then,

have an adaptive value zero, and will be opposed by a selection of

s = 1.00. What will be the frequency of tasters in the next genera-

tion? The homozygous tasters will contribute only T sex cells to the

gene pool of the next generation, whereas the heterozygous tasters

will produce equal numbers of T and t gametes. The gene frequen-

cies in the next generation will be 0.65 of T (0.204 + 0.247 = 0.451,

or 65 per cent of the total) and 0.35 of t (0.247, or 35 per cent of the

total).

Supposing that the non-tasters of PTC are prevented from reproduc-

tion generation after generation, the frequency of the t gene in the gene

pool of the population will decrease as shown in Table 6.1. An im-

Estimation of the Speed of Selection 121

TABLE ti.i

THE PROGRESS or SELECTION AGAINST A RECESSIVE GENE

(The table shows the frequency in the gene pool of a human population of the

gene for taste-blindness (t) expected if all taste-blind persons are prevented

from reproduction in several generations.)

Genera- Genera-

tions of Gene tions of Gene

Selection Frequency Selection Frequency

Initial 0.55 10 0.085

1 0.35 15 0.059

2 0.26 20 0.046

3 0.21 30 0.026

4 0.17 40 0.024

5 0.15 • 50 0.019

6 0.13 100 0.010

7 0.11 500 0.002

8 0.10 1000 0.001

9 0.09

portant conclusion to be gleaned from Table 6.1 is that selection

against a recessive gene is less effective the rarer that gene is in the

population. Two generations of selection suffice to reduce the fre-

quency of the gene t from the original 0.55 to less than half of the

original—0.26; but four generations are needed to cause another halv-

ing of the gene frequency, to 0.13. Starting with a frequency of 0.010,

we need 400 generations to depress the frequency down to 0.002.

This shows the immense difficulty which would be encountered by

selection against rare recessive defects or diseases in man, contem-

plated by some eugenical programs. Taking into consideration that the

length of a human generation is at present some 25 years, such eugeni-

cal measures would have very little effect in a predictable future.

In the example just considered, the selection was supposed to be

absolutely thorough (s = 1.0). With less intense selection, when only

a part of the recessives is debarred from reproduction (s less than 1.0),

the progress would be slower still. And yet, no matter how small the

selection may be, it will, given enough time, change the gene fre-

quencies in the direction favored by the environment or by the breeder.

This is important to know, because one of the objections made against

Darwin's theory of selection was that selection in nature will have no

effect since most variable traits of organisms are neither very useful

nor very harmful to their possessors. In reality, even small genetic ad-

vantages and disadvantages may be important in evolution.

122 Natural Selection and Adaptation

r

Adaptive Value and the Environment. Differences in the adaptive

value may be of diverse origin. Adaptive value, or Darwinian fitness,

is not the same thing as bodily strength, vigor, or bravery. The "fit-

test" may actually be a relative weakling if he gives birth to more

children than his hale and hearty neighbor.

Furthermore, the adaptive value of a genotype depends on the en-

vironment in which it lives, as examples given in Chapter 5 abundantly

show. Here we may consider only the situation in domesticated ani-

mals and plants and in man. Primitive cattle breeds yield only two

or three quarts of milk per cow per day; modern dairy breeds produce

several times more (as much as 10,000 pounds of milk and 300 or more

pounds of butterfat per cow per year). On the other hand, primitive

cattle can live well with little care under harsh conditions, whereas

their modern counterparts require special feeding and attention. Has,

then, the selection practiced for centuries by cattle breeders improved

or damaged the fitness of the domestic cattle? The question is evi-

dently meaningless if the environment is not specified. Clearly, the

breed which produces much milk has a superior fitness in the environ-

ment of a dairy farm, even though it becomes more fastidious in its

diet, less able to resist inclement weather, and quite unable to defend

itself against wolves and other enemies.

For a similar reason the often-met-with statement that natural selec-

tion has ceased to operate in modern human societies does not stand

critical examination. True enough, the importance for survival and

reproduction of such qualities as the ability to withstand cold with-

out protective clothing, to find and kill wild animals with bare hands

or with stones, to eat uncooked food, and to give birth without ob-

stetrical help is presumably less in modern man than in his remote

ancestors. The importance of possession of a nervous system which

permits learning complex techniques and withstanding the wear and

tear of modern "tempos" is presumably greater for modern man. Mod-

ern medicine has transformed some diseases that were fatal in primi-

tive man into passing annoyances in modern man, but modern life

requires a greater plasticity of behavioral traits than was needed be-

fore (cf. Chapter 14). In any case, whatever traits are favored o?

discriminated against under civilized conditions are by definition tht

traits which increase or diminish the fitness of their possessors in civil-

ized environments. Calling selection "natural" only if it selects traits

which we deem desirable is a lax use of words.

Modern theories of evolution consider the environment to be the

directive force in the evolutionary process. But the environment does

Asexual and Self-fertilizing Organisms 123

not change the organism from without, as some theorists of the past

believed. The environment furnishes the challenge, to which a living

species may respond by adaptive transformations of its gene pool.

Whether the species does or does not respond to the challenge de- ,

pends on the presence in the gene pool of the proper raw materials, I

mutant genes and gene combinations. Thus a culture of colon bac-

teria exposed to a bacteriophage produces a bacteriophage-resistant

strain only if it happens to contain resistant cells previously arisen by

mutation. No strains of colon bacteria capable of growing at a tem-

perature of 80 to 85° C have been found, and this species apparently

does not produce mutants which could resist such temperatures.

Nevertheless, algae which live at such temperatures are known in the

hot springs of Yellowstone Park.

Natural Selection and Self-reproduction. Natural selection is uni-

versal in the living world. It is implicit in the process of living. The

origin of life on earth meant the appearance of self-reproducing units

which tended to transform the susceptible materials (food) in the

environment into their own copies. However, the process of self-

reproduction is not absolutely perfect, and deviating, mutant, copies

appear from time to time. If the mutants reproduce themselves less

efficiently than the ancestral units, they are eliminated. If the mutants

are more efficient, they crowd out the ancestral form. Finally, if the

mutants can maintain themselves in an environment unsuitable to

the ancestor (that is, can subsist on a new food), the ancestral form

and the mutant may continue to exist side by side. Thus life is

diversified.

Self-reproduction brings in its train natural selection, and the two

together result in evolutionary progress. Self-reproduction, therefore,

may reasonably be regarded the fundamental property of life (Chap-

ter 1).

Selection in Asexual and Self-fertilizing Organisms. It will be

shown in Chapter 7 that "pure races" do not exist, and never have

existed, in man or in other organisms which reproduce sexually and

by cross-fertilization. But "pure races" can be formed in organisms

which propagate asexually, by parthenogenesis, or by self-fertilization.

In such organisms the entire offspring usually have the same genotype

as the mother. Unless mutation intervenes, clones and pure lines are

formed which consist of individuals with identical heredity. As shown

by Johannsen (cf. page 113), selection may be observed at its simplest

in organisms which form clones or pure lines. A field of wheat, or of

barley, or of beans, or a culture of bacteria usually consists of mix-

124 Natural Selection and Adaptation

tures of pure lines or clones. When a breeder picks out a seed from

the sturdiest plant in his field, he isolates a single pure line, which

may give rise to a new variety. Some of the most valuable varieties

of the self-fertilizing crop plants are descendants of single selected

individual progenitors. Within a pure line or a clone, selection is,

according to Johannsen, without effect, since the selected individuals

are not genetically different from the rest.

When a mixture of pure lines or clones is exposed to natural selec-

tion, the multiplication of some lines may be favored and other lines

may be discriminated against. Harlan and Martini planted the same

mixture consisting of equal parts of eleven varieties of barley in several

regions of the United States having different climates and other en-

vironmental conditions. For several years thereafter the mixture was

harvested and replanted again in the same regions. The proportions

of the varieties in the mixture have become radically altered: some

varieties were eliminated altogether, others increased in frequencies.

Interestingly enough, different varieties were favored by selection in

different regions. Thus the variety Hannchen was victorious in Minne-

sota, the variety White Smyrna in Montana and Oregon, Manchuria

in New York State, etc. (see Table 6.2).

TABLE 6.2

SURVIVAL OF BARLEY VARIETIES FROM THE SAME MIXTURE AFTER REPEATED

SOWINGS IN DIFFERENT ENVIRONMENTS

(After Harlan, Martini, and Stebbins.)

Vir- New Minne- Cali-

Varieties ginia York sota Montana Oregon fornia

Coast &Trebi 89.2 11.4 16.6 17.4 1.2 72.4

Hannchen 0.8 6.8 61.0 3.8 0.8 6.8

White Smyrna 0.8 0 0.8 48.2 97.8 13.0

Manchuria 0.2 08.6 0.4 4.2 0 0

Calami 2.6 1.8 3.2 11.6 0 0.2

Meloy 0.8 0 0 0.8 0 5.2

Others 5.6 11.4 18.0 14.0 0.2 2.4

Note. The figures show the percentages of the different varietiesiwhich re-

sulted from selection.

When a single "pure race" is isolated the progress of selection comes

to an end. However, the occurrence of a mutation in a pure line or

a clone may open a new field of action for selection, just as the occur-

rence of a mutation to streptomycin resistance in bacteria permits

Selection in Sexual Cross-fertilizing Populations 125

selection of a resistant strain (Chapter 5). The variety of oranges

known as "Navel" arose by mutation on a single tree found at Bahia,

Brazil, in 1870. The many thousands of Navel oranges cultivated in

California and elsewhere are direct descendants of the original mu-

tant propagated by grafting (that is, asexually). The smooth-skinned

peaches (nectarines) and some varieties of apples and other fruit

trees are descendants of single "lucky" mutants.

Selection in Sexual Cross-fertilizing Populations. The situation in

organisms which reproduce sexually and by cross-fertilization is very

different from that in species which form clones or pure lines. The

chief biological consequence of sexuality is emergence of an endless

variety of genotypes in sexual populations. As shown in Chapter 2,

an individual heterozygous for n genes is able to produce 2" kinds of

sex cells carrying different combinations of the parental genes. The

result is that in sexually reproducing species no two individuals are

likely to have the same genotype.

A relatively more constant characteristic of a sexual population is

its gene pool, in which various genes are represented, each with a

definite frequency. Except for identical twins, no two persons are

likely to have the same genotype, but every human population can

be described -as having certain frequencies of the blood group genes,

of genes for tasting and non-tasting certain chemical substances, for

different eye colors, skin colors, etc. (see Chapter 7 for further discus-

sion of this point). Natural or artificial selection may increase the

frequencies of some genes, and decrease the frequencies of others, in

the gene pool of a sexual population. Since new genotypes are con-

stantly produced by the sexual process, selection has always new ma-

terials to work with. The "improvements" brought about by selection

in sexual populations may, therefore, far exceed the limits of variation

in the original materials, even if no mutation intervenes. Surely, no

wild hen ever laid as many eggs as do modern poultry breeds, and no

wild boar reached weights which are usual for hogs on Iowa farms.

Long-continued application of artificial selection has brought about

radical changes in the genotypes of domesticated animals.

Table 6.3 shows the results of a selection experiment carried by corn

breeders and geneticists at the Illinois Agricultural Experiment Station

for more than half a century. The original variety with which the

work was started in 1896 had an average protein content of 10.92 per

cent and an average oil content of 4.70 per cent in its seed. Selection

was carried both in the direction of increasing and in the direction of

decreasing the protein and oil contents. By 1949 lines were obtained

126 Natural Selection and Adaptation

TABLE 6.3

PROGRESS OF SELECTION FOR HIGH AND FOR Low CONTENTS OF PROTEIN AND

OF OIL IN CORN

(After Woodworth, Lang, and Jugenheimer.)

High Lines Low Lines

Per Cent Per Cent Per Cent Per Cent

Generation Protein Oil Protein Oil

Initial 10.9 4.7 10.9 4.7

5 13.8 6.2 9.6 8.4

10 14.3 7.4 8.6 2.7

15 13.8 7.5 7.9 2.1

20 15.7 8.5 8.7 2.1

S5 16.7 9.9 9.1 1.7

30 18.2 10.2 6.5 1.4

35 20.1 11.8 7.1 1.2

40 22.9 10.2 8.0 1.2

45 17.8 13.7 5.8 1.0

50 19.4 15.4 4.9 1.0

having as much as 19.45 per cent protein and 15.36 per cent oil, and

lines with as little as 4.91 per cent protein and 1.01 per cent oil. It

does not seem that the selection has yet reached the limits of its

effectiveness.

Selection and Hybridization. The success of selection depends

upon the presence of genetic variability in the population to which

the selection is applied. As we have seen, among asexual or self-

fertilizing organisms selection can only isolate the best genotypes

already present when the selection is started. Sexual crossing and

recombination of genes among hybrids yield a vast supply of ever-

new genotypes, some of which may be superior to the old ones. This

is the reason why many valuable varieties obtained by modern breed-

ers are isolated from among hybrids between two or several older

varieties. For example, the Thatcher wheat variety which now grows

on many millions of acres of wheat fields has a combination of genes

derived from the varieties lumillo, Marquis, and Kanred.

In sexual organisms hybridization is, in a sense, the regular method

of reproduction. Any heterozygous genotype is a hybrid genotype.

In this sense, any human being is a hybrid, and any marriage a hy-

bridization. But the word "hybridization" is also used to describe

crossing of more or less dissimilar parents: different breeds, races, and

species. The more dissimilar are the parents crossed, the greater is

Genetic Drift 127

likely to be the number of differences between them, and the greater

the diversity of genotypes resulting from segregation in hybrid prog-

enies. Breeders of domestic animals and plants often resort to more

or less remote hybridization. Thus the 'Thoroughbred" horse arose

several centuries ago from hybrids between the light Arabian and

Barbary, and the heavier European horses (see Chapter 9). To take

a more recent example, the Santa Gertrudis cattle are derivatives of a

cross between shorthorn and the Indian (zebu) cattle.

Introgressive Hybridization. In making use of hybrids between

dissimilar varieties and species, the breeders have borrowed a method

which is of some importance also in evolution in nature, at least in the

plant kingdom. The work of botanists (in recent years particularly

of Anderson and of Stebbins) has shown that distinct species of plants,

when they inhabit the same territory, occasionally cross and produce

fertile offspring. Of course crossing within a species remains more

frequent than that between species. Interspecific crosses, however,

cause a diffusion, or introgression, of genes from one species to an-

other. The gene transfer augments the genetic variability and fur-

nishes great opportunities for natural and artificial selection.

Iris fulva and Iris hexagona are two species of irises studied by

Riley in southern Louisiana. They differ in a number of ways: in the

former species the tube of the flower is golden yellow, and in the

latter it is green; in the former the sepals are coppery red and rela-

tively short, in the latter they are blue-violet and relatively long, etc.

Furthermore, Iris fulva grows chiefly on drier lands, in woods, and on

slopes of ridges; 7ris hexagona occurs on low lands, on wet clay, and

in full sun. It appears that before man came and started to cut the

forests and to drain the marshes for pastures and fields, the two

species of irises kept well separated in their respectively different

habitats. Man's activities have created new habitats on which both

species, as well as hybrids between them, can grow. Riley has found

colonies of irises in which varying proportions of individuals were

clearly of hybrid origin, containing mixtures of genes of both species.

Another example of hybridization, involving the cultivated corn

(maize), will be discussed in Chapter 9.

Genetic Drift. Another agency which cooperates with natural selec-

tion in bringing about evolutionary changes is genetic drift, the action

of which was studied mathematically in a series of important works

chiefly by Sewall Wright. The nature of this agency is simple enough.

Since a sex cell contains only half the chromosomes present in the

diploid organism, a parent transmits to each child only half of his

128 Natural Selection and Adaptation

genes, and fails to transmit the other half. In a stationary population

(that is, a population which neither expands nor contracts in num-

bers in successive generations) every pair of parents produces on the

average two surviving offspring. With two children, half of the

parental genes will be represented once each, one quarter will be

represented twice each, and one quarter will be altogether lost. In

the next generation some of the genes will again be reduplicated and

others will be lost. In the course of many generations some "lucky"

genes will be considerably increased in frequency, but many others

will no longer exist. For this reason, it can be said that only a fraction

of the genes which were present in the gene pool of the human species

in, for example, the time when the pyramids were being built, or dur-

ing the Stone Age, still exist in the now-living mankind. But these

genes have become, barring mutation, greatly increased in numbers.

In reality, even in a stationary population the offspring of some

parents are lost, whereas other parents leave numerous progeny. Many

populations (of, for example, insects) alternately expand greatly in

numbers in some years, and then are decimated until only few indi-

viduals are left. The genetic consequence of such expansions and con-

tractions is that many genes are lost and others are increasing in

frequencies. The author once visited a village with a population of

perhaps three hundred individuals isolated in a remote corner of Asia.

The singular condition in this village was that most of the inhabitants

had the same family name. The explanation of this condition was

simple—apparently an early settler happened to raise a large family

consisting mostly of boys, and his sons were in turn fathers of large

families. The genes of the early settler have become, within perhaps

half a dozen generations, incorporated in many individuals. If this

settler lived in a large city, the multiplication of his family name would

not be conspicuous, since the carriers of this name would be scattered

among other people. In a village the result seemed striking. Sup-

pose, then, that the early settler carried a gene or a chromosome struc-

ture which is rare in the general population. The village in question

has come to have a population in the gene pool of which this gene or

chromosome could be unusually frequent.

As pointed out on page 119, the Hardy-Weinberg theorem of con-

stancy of gene frequencies presupposes that the population consists

of infinitely many breeding individuals. The individuals of a given

generation may be said to come from a random sample of sex cells

taken from the gene pool of the preceding generation. Suppose that

this gene pool had equal numbers of the genes A and a, q = (1 — q)

Drift and Selection in Isolated Populations 129

= 0.5. Suppose, next, that the sample consists of one million sex cells

(500,000 individuals). The gene frequency in the next generation

may deviate from 0.5 by a standard error which will be equal to

V0.5 X 0.5 : 1,000,000 = 0.0005. This standard error is only one-

thousandth of the gene frequency, and, consequently, the gene fre-

quency may be said to be unchanged. The situation will be different

in a sample of 100 sex cells (50 individuals). The standard error will

now be V0.5 X 0.5 : 100 = 0.05. This is one-tenth of the original

gene frequency, and the gene frequency in the next generation will

probably lie between 0.45 and 0.55, but it may easily be between

0.4 and 0.6, and even as low as 0.35 or as high as 0.65.

Variations in gene frequencies due to sampling errors in finite popu-

lations are known as genetic drift. Mutation and selection tend to

change the gene frequency in a population systematically in the same

direction, generation after generation. Genetic drift has an unsys-

tematic effect: in one generation it may pull the gene frequency up-

ward, then downward, or leave it unchanged. Nevertheless, given

enough time (enough generations), the genetic drift may produce large

changes in populations consisting of small numbers of breeding in-

dividuals.

Drift and Selection in Isolated Populations. Many organisms do

not occur everywhere in the countries in which they live, but form

more or less isolated groups, or colonies, each containing only some

tens or hundreds of individuals. Thus land animals and plants may

form colonies on dry land isolated by water, whereas water-dwelling

organisms often live in lakes or ponds isolated by stretches of land.

Many creatures occur only on certain soils, or live only where a cer-

tain kind of food is available. At the dawn of history mankind con-

sisted of many small nations or tribes, more or less isolated from each

other, with marriages occurring mostly within a tribe.

Suppose that the isolated colonies of a species have, to begin with,

similar gene pools, with a certain gene A having the frequency, for

example, of 0.5. If this gene is not acted upon by mutation and selec-

tion, its frequency in the species as a whole will continue to remain

0.5, in accordance with the Hardy-Weinberg rule. But consider the

separate tribes. If these tribes have small populations, the gene fre-

quencies will, owing to genetic drift, become higher than 0.5 in some

tribes and lower than 0.5 in others. In the course of time, the tribes

will tend to drift apart in gene frequencies. In fact, some tribes may

come to have only the gene A and others only a.

130 Natural Selection and Adaptation

It is possible, although not proven (cf. Chapter 7), that the differ-

ences between human races in traits which seem to be adaptively

neutral arose through genetic drift in small populations. Most human

populations contain persons of different blood groups, tasters and non-

tasters of PTC, etc.; however, the incidence of these traits varies in

different parts of the world (Figures 7.6 and 7.7). These variations

could have arisen through genetic drift in the isolated tribes of which

early mankind was composed, and subsequently impressed upon the

much larger and no longer strictly isolated modern races. Colonies

and local races of many animals and plants often show minor differ-

ences in color patterns, size, shapes of various parts, etc., which also

may be neither useful nor harmful to their possessors. Genetic drift

in small populations is one of the possible hypotheses to account for

such differences between populations.

Interesting situations arise in small populations which are acted

upon simultaneously by natural selection and by genetic drift. Selec-

tion always tends to fix the frequencies of genes in the populations at

values most advantageous under the environments in which these

populations live. The genetic drift causes variations up and down

from these values. The separate, partially or completely isolated,

populations of a species act, then, as evolutionary trial parties which

explore the adaptive possibilities of various genetic structures. Most

of these trial parties "discover" nothing of interest, and eventually get

lost and supplanted by the more successful populations. The impor-

tant thing, however, is that in some populations, perhaps in only a

single population, gene combinations may arise which are not very

useful, perhaps even slightly disadvantageous, but possess new adap-

tive potentialities. Such "evolutionary inventions" are then perfected

by natural selection, and occupy new ecological niches which were

unattainable to the original species. The importance of the genetic

drift in evolution is that its interaction with natural selection confers

a greater adaptive plasticity than a living species would possess other-

wise.

Conservative and Creative Aspects of Selection. Some philosophers

and some biologists have doubted that natural selection can be the

guiding agent in evolution because selection, allegedly, produces

nothing new, and merely removes from the populations degenerate

variants and malformations (cf. Chapter 14). This objection seemed

particularly impressive when Johannsen demonstrated that selection

does not induce hereditary variations in genotypically uniform pure

Conservative and Creative Aspects of Selection 131

lines and clones (see page 114), but the objection became invalid in

the light of modern biological knowledge.

We should clearly distinguish the two basic evolutionary processes:

that of the origin of the raw materials from which evolutionary changes

can be constructed, and that of building and perfecting the organic

form and function. Evolution can be compared to a factory: any fac-

tory needs a supply of raw materials to work with, but when the

materials are available they must be transformed into a finished

product by means of some manufacturing process. The raw materials

of evolution are the genotypes which arise by gene and chromosome

mutation and by recombination engendered in the process of sexual

reproduction. But, according to the apt phrase of Sewall Wright,

mutation alone would produce "an array of freaks, not evolution." It

is selection which gives order and shape to the genetic variability,

and directs it into adaptive channels. Some of the classical theorists

of evolution believed that evolution was produced by natural selection;

others thought that it was produced by mutation or by hybridization.

They were equally in error, since it is the interaction of all these

processes which results in evolutionary changes.

Mutation and recombination yield a supply of genetic variability,

which is then organized by natural selection in accordance with the

demands of the environment. Most mutants, however, are deleterious

to the organism in "normal" environments; they produce defects, mal-

formations, and hereditary diseases of various degrees of gravity. The

action of selection, accordingly, has two aspects, which Schmalhausen

has called the stabilizing selection and the dynamic selection. The

former keeps down the number of deleterious mutant genes and gene

combinations (cf. Chapter 7), and in so doing it protects and stabilizes

the "normal" developmental pattern of the species. On the other hand,

the environment changes in time and in space. The dynamic natural

selection permits the species to keep its hold on the changing ecologi-

cal niches which it occupies, and also to conquer and control new

ecological opportunities.

Even the stabilizing function of natural selection involves more than

a purely negative action of blocking the spread in populations of bad

mutants. It is frequently forgotten that the environment which we

call "normal" for a species is in reality a complex of a great many dif-

ferent environments. Just think how profoundly different are the

environments to which any species inhabiting temperate or cold lands

is exposed in summer and in winter. In order to survive, such species

must of necessity be adapted to all the environments which recur

132 Natural Selection and Adaptation

seasonally, or from time to time, in their native country. Many trop-

ical and subtropical plants can grow well enough during the summers

in the United States, but they do not survive winters; in California

and Florida some of them survive average winters but are killed by

exceptionally severe ones. How many, if any, offspring are produced

by the carriers of any genotype depends on the reactions of this

genotype not to just one but to many different environments—in other

words, on the norm of reaction of this genotype (cf. Chapter 4).

Adaptively most valuable genotypes are those which react favorably

in all the environments to which the species is exposed in its natural

habitats. The stabilizing natural selection favors, then, the genotypes

which can withstand the environmental shocks which the organism is

likely to meet. The result of selection is the development of homeo-

stasis (cf. Chapter 1). The organism of man, of a frog, of a Drosophila

fly, of a corn plant, and in fact of any species develops and maintains

its normal structures and functions despite the manifold and often

erratic changes in the environments. The "stabilization" of organic

development permits a great deal of genetic and evolutionary progress.

New developmental patterns become necessary, however, when the

environment changes permanently. This change may occur because

of alterations in the climate, because of ecological upsets produced

by the appearance of new diseases, parasites, predators, or competi-

tors, because the species becomes adapted to eat new kinds of food,

to grow on different soils, and for many other reasons. Dynamic

natural selection involves, then, the establishment of new genotypes,

adapted to the new conditions. It is fair to say that selection produces

new genotypes, even though we know that the immediate causes of

the origin of all genotypes are mutation and recombination. It is only

in organisms which reproduce exclusively by asexual means or by self-

fertilization that selection, in the short run, merely picks out some of

the available genotypes and suppresses others. In the long run, selec-

tion is the directing agent because it determines which genotypes are

available for new mutations to occur in.

Sexual reproduction and cross-fertilization are immensely efficient

trial-and-error mechanisms which generate new genotypes. Their ac-

tion in a given generation depends, however, on the composition of

the gene pool of the population, which in turn has been determined

through natural selection in the ancestral generations. Thus it comes

about that we can obtain, by selection, strains with characters which

far exceed the limits of variation in the ancestral materials (see page

125). The human species, and all other species of organisms which

Suggestions for Further Reading 133

live in the world at present, were not preformed in the primeval amoe-

bae or in the primordial viruses. They have evolved gradually in the

history of the earth under the control of natural selection.

Suggestions for Further Reading

Darwin, Ch. 1859. On the Origin of Species by Means of Natural Selection.

This is the original statement of the theory of natural selection. For further

comment, see the "Suggestions for Further Reading" at the end of Chapter 5.

Osborn, H. F. 1896. From Greeks to Darwin. 2nd Edition. Macmillan, New

York.

Although antiquated, this work remains a useful account of the early history

of evolutionism.

Andrewartha, H. G., and Birch, L. C. 1954. The Distribution and Abundance

of Animals. Chicago University Press, Chicago.

Dobzhansky, Th. 1951. Genetics and the Origin of Species. 3rd Edition. Co-

lumbia University Press, New York.

Schmalhausen, I. I. 1949. Factors of Evolution, the Theory of Stabilizing

Selection. Blakiston, Philadelphia.

Stebbins, G. L. 1950. Variation and Evolution in Plants. Columbia University

Press, New York.

These books (Andrewartha and Birch to Stebbins) should be consulted for dis-

cussions of modern versions of the theory of natural selection.

Allce, W. C. 1951. Cooperation among Animals. Schumann, New York.

Kropotkin, P. 1917. Mutual Aid. Knopf, New York.

Also available in a Penguin Books edition (London, 1939).

Ashley Montagu, M. F. 1952. Darwin, Competition, and Cooperation.

These three books (Allce to Ashley-Montagu) emphasize cooperation, as against

struggle, as an important factor leading to natural selection.

Li, C. C. 1955. Population Genetics. Chicago University Press.

Lerner, I. M. 1950. Population Genetics and Animal Improvement. Cambridge

University Press, London.

The books of Li and Lenier may be consulted for a mathematical theory of

selection and of genetic drift. These books are intended for advanced students

and research workers.

For a very elementary treatment of the same topics see also:

Dunn, L. C., and Dobzhansky, Th. 1952. Heredity, Race, and Society. 2nd

Edition. New American Library, New York.

Anderson, E. 1949. Introgressive Hybridization. John Wiley, New York.

The author of this last book believes that introgressive hybridization is rather

more important in the evolution of life than it is usually credited with being.

Another discussion of this phenomenon can be found in the book by Stebbins

quoted above.

7

Individuals, Populations,

and Races

Experience shows that every person we meet is different from any

met before. Individual differences exist also among animals and

plants, and for that matter no two material objects are completely

identical. Yet human language forces the infinite variety of experi-

ence into categories symbolized by words—man, horse, dog, pine, etc.

It is easy to mistake words for actual objects, and to conclude that

each word refers to some metaphysical entity or "idea." Plato, the

greatest philosopher of antiquity, actually taught that individual men,

horses, pine trees, etc., are imperfect and temporary expressions of

the eternal and unchangeable ideal Man, Horse, and Pine.

Although few modern philosophers and still fewer scientists take

Plato's "ideas" literally, his way of thinking is deeply rooted in many

minds. It is common to hear people speak glibly of a "typical French-

man," or "real American," or "ideal horse." Such expressions are

legitimate only so long as the speaker realizes that the "type" or

"ideal" is a composite image which he endows with properties com-

monly met with in actual individuals or considered desirable or pleas-

ing. The trouble is that people are frequently tempted to think of

these abstractions as though they were real entities. This "typological"

thinking may even be carried to the point when the imaginary Man

or the imaginary American is substituted for real men and for living

persons who compose a nation as objects of sympathy and affection.

Types and Classifications. The typological approach is convenient

in the branches of biology which deal with description and classifi-

cation of animals and plants. Just as a large library must be systema-

tized and catalogued to be usable, so living organisms have been di-

vided into phyla, classes, orders, families, genera, species, and sub-

134

Pure Races? 135

species or races, and each category has been given a name. A name

stands for a group of individuals which have some traits in common.

Thus all individuals called "fox terrier" have many properties in com-

mon; some of these properties are shared by all individuals referred

to the species dog (Cants familiaris), to the genus Canis, the family

Canidae, the order of carnivores (Carnivora), the class of mammals

(Mammalia), and the phylum of vertebrates (Chordata). The prop-

erties common to each category may be represented schematically in

a picture or in a description; such schemes are very useful in teaching

zoology and botany.

The "types" of dog, carnivores, vertebrates, etc., do not exist, how-

ever, apart from the real animals. It was perhaps this kind of illusion

which led Linnaeus to conclude that species were units created by

God. In the nineteenth century the French comparative anatomist

Cuvier and many others believed in types or "basic plans" which

existed in some ideal world and of which the real animals and plants

were but imperfect copies. The great poet Goethe created a schematic

generalized plant ("Urpflanze"), of which the plants that actually

grow were supposed to be variants (Chapter 10).

Much harm came from the notions of typical or "pure" white,

Nordic, Germanic, and other "pure" races of man. No two men look

alike, and different countries are inhabited by people who look

obviously different. Some anthropologists, from the eighteenth cen-

tury to the present time, have succumbed to the temptation of sup-

posing that this great variety of human beings arose through mixture

of a relatively small number of "pure" races. These anthropologists

assume that the "pure" races have lived at some unspecified time in

the past, and arbitrarily endow them with combinations of traits, such

as stature, head shape, skin, and hair colors. Very occasionally a living

person is found who more or less resembles the imaginary standard of

the "pure" race, but a great majority of human beings have to be

regarded as mixtures of two or of several races. Authorities seldom

agree on either the number or the characteristics of the supposed

"pure" races; this disagreement is not surprising since, in the words

of the eminent anthropologist Howells, these races were arrived at by

"a kind of divination." All this would have made trouble only for

anthropologists, but politicians and bigots found the "pure" races

convenient instruments to stir up race prejudice and unhealthy na-

tionalism.

Pure Races? As shown in the foregoing chapter, pure races con-

forming to a single genetic type could exist in sexually reproducing

136 Individuals, Populations, and Races

organisms only if heredity were transmitted by blood instead of by

genes. Then the "type" of a race or of a species could represent the

limiting condition which a sexually reproducing population would

reach after a prolonged period of breeding in isolation. But in reality

groups of genetically uniform individuals—clones and pure lines—exist

only in asexual or self-fertilizing populations. Hardy and Weinberg

(see Chapter 6) have firmly established that sexually reproducing and

cross-fertilizing populations do not become any "purer" with time, and

that when applied to such populations the notion of "pure" race is

absurd.

Although members of a sexual cross-fertilizing population are rarely,

if ever, genetically identical, and although individuals of a clone or a

pure line usually are genetically uniform, individuals in a sexual species

are far more interdependent than those in clones or pure lines. In a

clone of bacteria, every cell may give rise by division to a progeny,

but individuals of a clone are independent from one another in repro-

duction. In sexual species, an isolated individual can leave no progeny

and is lost to the species. Members of a sexual population are inter-

dependent. They are bound together by ties of mating and parentage.

Every individual derives its genes from the common gene pool of the

population, and returns its genes to the same gene pool.

The word "population" is sometimes used loosely to refer to any

group of living beings; thus we speak of the bird population of a

forest, or of the fish population of a lake. But a Mendelian population

is a reproductive community of sexual and cross-fertilized individuals

among whom matings regularly occur and who, consequently, have a

common gene pool.

In theory, Mendelian populations may be homozygous for all their

genes, but this happens rarely or never in reality. Every member of a

Mendelian population (identical twins excepted) is likely to have

a genotype of its own, not found in any other individual. It would

be meaningless to say that every individual belongs to a separate

"race." It is also easy to see how little meaning, except in a purely

descriptive sense, has the idea of "type" when applied to a Mendelian

population. Such expressions as the "average" condition or the "typi-

cal" phenotype may be entirely misleading. For example, most human

populations consist of persons with O, A, B, and AB blood groups, but

there is no such thing as an "average" blood group. We can say only

that in some populations persons belonging to certain blood groups are

relatively more frequent than those belonging to other blood groups.

Sympatric and Allopatric Variability 137

Such a statement does not, however, describe an average or a type,

but the composition of the gene pool of the population.

Sympatric and Allopatric Variability. Mendelian populations may

be of various orders of magnitude. In man marriages are concluded

most often among residents of the same community, town, city, and

country. Every individual human being encounters during his repro-

ductive age a certain number of individuals of the opposite sex who

may be regarded as his potential mates. These groups of potential

mates are the elementary Mendelian populations, to which in human

genetics and anthropology the term isolates is often applied. But the

isolates are seldom completely isolated from each other. Intermar-

riages between isolates occur more or less frequently, giving rise to

Mendelian populations of higher orders. The language which people

speak, the nation and the country to which they belong, influence the

frequencies of intermarriage and make larger isolates. Finally, man-

kind as a whole is a great Mendelian population. Any two human

beings of reproductive age and of opposite sex anywhere in the world

are potentially able to mate and to produce offspring. There is ac-

tually some gene flow, direct or indirect, continuous or intermittent,

between all isolates of the human species.

Individuals of the same or of different species who live together in

the same territory are called sympatric. Residents of different terri-

tories are allopatric. These terms may be defined more precisely as

follows. The places where the parents are born are usually separated

by some distance from the birthplaces of their offspring. Individuals

who live within the average distance between the birthplaces of par-

ents and offspring are sympatric, and those living at greater distances

are allopatric. Matings occur usually between sympatric individuals,

and these individuals are members of elementary Mendelian popula-

tions. Although this distinction is obviously not sharp, because ad-

jacent populations commonly overlap, it is of considerable importance

in evolutionary biology.

Genetic differences are observed, of course, between sympatric as

well as between allopatric individuals. People in an American town

differ in appearance from the inhabitants of a Japanese town, but

individual Americans as well as individual Japanese differ also among

themselves. But the differences between sympatric individuals arise

through combination of genes coming from the same gene pool, and,

unless the individual dies childless, the genes revert again to the

gene pool. Allopatric populations have separate gene pools, although

there may be some diffusion of genes between these pools, owing to

138 Individuals, Populations, and Races

migration and to intermarriage of people whose parents were born in

remote parts of the world.

The existence of genetic differences between sympatric individuals,

members of the same Mendelian population, is referred to as individual

variability or as polymorphism. Genetic differences between Mendel-

ian populations are called racial or subspecific. Race and subspecies

are synonyms, the former word being used most often in connection

with man, and the latter for organisms other than man. Breeds of

domestic animals and varieties of cross-fertilized cultivated plants are

races. It should be noted, however, that, although most races of free-

living species are allopatric, breeds and varieties of domesticated

species may be either sympatric or allopatric, since their mating is con-

trolled by man. Human races were originally allopatric, but with the

development of the cultural regulation of marriage they have become

partly sympatric (see Chapter 13). The word "variety" has been

used in so many different senses that it is at present worthless as a

scientific term. It is sometimes applied to clones of asexual species

and even to sympatric genotypes within a polymorphic Mendelian

population, which cannot legitimately be called races or subspecies.

Some Examples of Polymorphism in Human Populations. As shown

in Chapter 6, a Mendelian population may be described most ade-

quately in terms of relative frequencies of various genes and chromo-

somal variants in its gene pool. For example, most human populations

are polymorphic with respect to the blood groups. The American

white population consists of approximately 47 per cent of persons with

blood of group O, 43 per cent of group A, 7 per cent of group B, and

3 per cent of group AB. Since human populations are panmictic with

respect to the blood groups (nobody chooses his mate according to

the blood group), it can be deduced that the three alleles which de-

termine the blood groups, 7°, 7A, and Is, have the frequencies in the

gene pool respectively of 0.67, 0.26, and 0.07.

Of course human populations are polymorphic with respect to many

genes other than those which determine the blood groups. In the

foregoing chapter it has been mentioned that some 70 per cent of

American whites are tasters, and some 30 per cent are non-tasters, of

the PTC substance. In the gene pool, about 55 per cent of the sex

cells carry the recessive gene t for non-tasting, and the remaining 45

per cent contain the dominant gene T for tasting.

The phenotypic manifestation of the genetic polymorphism in Men-

delian populations may or may not be very conspicuous. In human

populations the polymorphism with respect to the blood groups is

Recessive Deleterious Genes 139

perfectly clear-cut, since, when tested by competent observers, there

is never any doubt about the blood group to which a person belongs.

The situation with tasting the PTC substance is a little less definite,

since some persons find weak solutions of PTC tasteless but strong

solutions bitter. Classification of such persons as "tasters" or "non-

tasters" may be ambiguous. With many other traits, especially those

determined by polygenes (cf. Chapter 2), the different genotypes fail

to produce discrete phenotypic classes. Thus the pigmentation of

human skin varies all the way from albino to black. Similarly, the

color of the hair and eyes, the shape of the head, and countless other

human characteristics vary continuously from one extreme to another.

Dominant Deleterious Genes and Hereditary Diseases. We have

seen in Chapters 4 and 5 that, in any living species, there occur from

time to time mutations and chromosomal changes, most of which are

more or less deleterious to their carriers in the environments which

are normal for the species. There exist, of course, all transitions

between mutants which produce dangerous hereditary diseases and

mutant genes which are neutral and even useful to the organism.

Some mutants act as dominant lethals, that is, they kill their carriers

even in heterozygous condition. Thus in Drosophila many eggs de-

posited by an untreated female mated to a male treated with X-rays

fail to produce larvae. The death of the eggs is due to induction by

X-rays of dominant lethals (mostly inviable chromosome breakages)

in the treated spermatozoa. An example of a spontaneously arising

lethal in man is retinoblastoma, a form of tumor in the retina of the

eye, which causes death usually during infancy. According to Neel

and Falls, about 4 in 100,000 infants born in the State of Michigan

are new retinoblastoma mutants.

Deleterious mutants which are not completely lethal may be re-

tained in a population for several generations. This is the case with

numerous hereditary defects and diseases in human populations. Thus

the chondrodystrophic dwarfism (see Chapter 2, page 32) is a semi-

lethal in the genetic sense. These dwarfs enjoy satisfactory health,

but, according to Haldane, produce on the average only about one-

fifth as many children as do normal persons.

Recessive Deleterious Genes and Hereditary Diseases. A newly

arisen recessive mutant is likely to be carried in an individual heterozy-

gous for the recessive and for the dominant normal allele. Even if the

recessive mutant is lethal when homozygous, the heterozygous carriers

may be normal in appearance and in health. A vast majority of genes

for rare recessive hereditary diseases are carried in populations in

140 Individuals, Populations, and Races

normal-appearing heterozygotes, making it difficult to detect and to

eliminate them by any plausible eugenical measures. Thus the

juvenile amaurotic idiocy causes the death of the homozygotes before

the onset of sexual maturity, and yet the parents of the afflicted chil-

dren are normal people. Albinism in man may be classed as a mildly

deleterious condition due to homozygosis for a recessive gene, but the

parents of the albino are normally pigmented. Muller has estimated

that a "normal" and healthy person in human populations is heterozy-

gous, on the average, for eight recessive genes which could produce,

in homozygous condition, more or less grave defects or diseases.

Perhaps most interesting at this point are human genes which are

disabling or even fatal to homozygotes but are also detectable in

heterozygotes. Individuals homozygous for the gene for thalassemia

die of a fatal anemia, usually in childhood. The parents of the anemics

may have a mild form of anemia ("thalassemia minor"), and their red

blood cells show certain peculiarities detectable under the microscope.

The gene is, then, lethal in homozygous and subvital in heterozygous

condition. The remarkable condition is that the thalassemia occurs

chiefly among people native to the Mediterranean region (Italians,

Greeks, Syrians, etc.). In populations of some parts of Italy this gene

reaches, however, astonishingly high frequencies—10 or more per cent

of the gene pool.

Another form of lethal anemia is due to homozygosis for another

recessive gene which occurs chiefly among people of African extrac-

tion. This anemia, or sickle-cell disease, causes the red blood cells to

assume characteristic shapes (Figure 7.1) when placed in a medium

deficient in oxygen. The heterozygous carriers of this gene are normal

or very mildly anemic. The frequency of the sickle-cell gene in some

Negro populations exceeds 40 per cent, although in other populations

it is low.

It is certain that man is not the only species afflicted by accumula-

tion of deleterious genes in its gene pool. Most, or all, sexually repro-

ducing organisms share the same fate. The most detailed relevant

studies, begun in 1925 by a Russian geneticist Chetverikov, have been

made on natural populations of Drosophila flies. It may be noted that

a fair number of mutants had been found before 1925 in Drosophila

in laboratory cultures, but the outdoor populations of these flies

seemed quite uniform and free of mutants. So much so, that some

authorities at that time suspected that mutants are just laboratory

artifacts which never occur in nature.

142 Individuals, Populations, and Races

TABLE 7.1

PERCENTAGES OF THE SECOND AND THIRD CHROMOSOMES IN SOUTH AMERICAN

POPULATIONS OF Drosophila wUlistoni WHICH CONTAIN VARIOUS DELETERIOUS

RECESSIVE MUTANT GENES

(After Pavan and Collaborators.)

Second Third

Effects in Homozygotes Chromosome Chromosome

Lethal or semilethal 41.2 32.1

Subvital 57.5 49.1

Sterility 31.0 , 27.7

Retarded development 31.8 35.7

Accelerated development 13.7 2.8

Visible abnormalities 15.9 16.1

other hand, we must remember that Table 7.1 shows the total fre-

quencies of all kinds of lethals, semilethals, subvitals (relatively mild

hereditary diseases), sterility genes, etc. Most of these mutant genes

are not allelic, and flies heterozygous for two or even for several lethal

genes are quite viable and "normal" in appearance.

Heterosis and Balanced Polymorphism. Most of the genes consid-

ered up to this point produced heterozygotes which were similar to, or

intermediate between, the corresponding homozygotes (that is, the

heterozygotes, Aa, either resemble one of the homozygotes, AA or aa,

or are intermediate between them). It happens, however, that some

genes produce heterozygotes, Aa, which are superior in vigor or in fit-

ness to both AA and aa homozygotes. A superiority of the hybrids,

heterozygotes, is known as hybrid vigor, or heterosis. This condition

is common in nature, and it is very important in agricultural practice.

The exploitation of heterosis in hybrid corn in the United States has

resulted in an increase of the yield of about 15 bushels per acre. It is

estimated that the aggregate increase of the yield for the whole coun-

try in 1946 was between 832 and 924 millions of bushels (see Chapter

9).

The dynamics of a population with heterotic heterozygotes can be

described as follows. Suppose that the adaptive value of a heterozy-

gote, Aa, is equal to 1. The adaptive values of the homozygotes, AA

and aa, are, respectively, 1 — «i and 1 — s<> (st and s2 are, of course,

selection coefficients, see Chapter 6). Suppose that the frequencies

of the genes A and a in the gene pool of the population are q and

(!—
value of a heterozygote is above those of both homozygotes, natural

Chromosomal Inversions in Drosophila Populations 143

selection will act to retain both alleles, A and a, in the population in-

definitely. The population will, accordingly, be polymorphic; poly-

morphism maintained by selection is known as balanced polymorphism.

The genes under the conditions outlined above will reach equilibrium

frequencies at the following values:

q = «2/(*i + *2) and (1 - q) = sl/(sl + s2)

This little exercise of mathematics leads to interesting biological

conclusions. First of all, the mechanism of balanced polymorphism

will maintain in the population any gene which produces a heterotic

heterozygote, no matter how poorly viable, or even lethal, may be the

homozygotes. Assume, for example, that the homozygote, aa, is lethal

(s2 = 1-0) and the other homozygote, AA, is slightly less fit than the

heterozygote (let si be 0.1). Natural selection will establish an equi-

librium at which the gene a will have a frequency of (1 — q) per

cent in the gene pool [0.1/(1 +0.1) = 0.09].

It is quite conceivable that the occurrence of some recessive heredi-

tary defects and diseases in human and other species may be kept

up by balanced polymorphism. There is some evidence (Allison,

1954) that heterozygous carriers of the gene for the sickle-cell anemia

are less susceptible to malaria (tropical fevers) than are the "normal"

homozygotes. In countries where malaria is prevalent even a relative

immunity may constitute an important adaptive advantage. Natural

selection will then maintain a certain incidence of the genes for the

sickle-cell anemia in the population of malaria-ridden countries, despite

the fact that the homozygotes for this gene die of fatal anemia.

Chromosomal Inversions in Drosophila Populations. Experimental

verification of the hypothesis of balanced polymorphism proved pos-

sible in populations of Drosophila. Natural populations of some spe-

cies of these flies vary with respect to a rather recondite trait, namely,

the structure of their chromosomes. The arrangement of the genes in

the chromosomes is visibly reflected in the giant chromosomes of the

salivary glands of Drosophila larvae (Figure 7.2). Different chromo-

somal variants found in the same free-living population differ in in-

versions of blocks of genes (cf. page 66). The flies which have

chromosomes with different gene arrangements interbreed freely;

therefore, some flies possess two chromosomes of a pair with identical

gene arrangements (inversion homozygotes), and other flies have the

two chromosomes with different gene arrangements (inversion het-

erozygotes).

146

Individuals, Populations, and Races

kinds of chromosomes occur side by side. Now, as shown above, such

an equilibrium is expected if the inversion heterozygotes are superior

in adaptive value to the homozygotes.

Figure 7.4 shows the outcome of an experiment in which a popula-

tion of Drosophila pseudoobscura placed initially in the cage had 10

1.00

0.90

0.80

0.70

» 0.60

g 0.50

I

it 0.40

0.30

0.20

0.10

IT

8 10 12

Generations

14

16

!

18

20

50 100 150 200

250

Days

300 350 400 450 500

Figure 7.4. Progress of natural selection in a population of Drosophila pseudo-

obscura kept in a population cage (see Figure 7.3). The gene pool of this popu-

lation contained originally about 11 per cent of one kind of chromosome and 89

per cent of another. The bars show a gradual increase in the frequency of the

chromosomes which were originally rare in the population, and an eventual

attainment of a stable equilibrium.

per cent of the chromosomes with the gene arrangement called "ST"

and 90 per cent of the chromosomes with the arrangement known as

"CH." The graph shows that the incidence of the ST chromosomes

in the population rose rapidly (and that of CH declined). However,

in about 250 days after the start of the population (corresponding to

some ten fly generations), the population reached an equilibrium

state, with approximately 70 per cent ST and 30 per cent CH chromo-

Units of Natural Selection 147

somes. The changes observed in this population indicate the follow-

ing adaptive values for the different chromosomal types:

Adaptive Selection

Chromosomal Types Value Coefficient

Heterozygotes KT/CH 1.00 0

Homozygotes .ST/S71 0.90 *i = 0.10

Homozygotes CH/CH 0.41 *2 = 0.59

According to the formula given above (page 143), the population

is expected to reach an equilibrium when the ST chromosomes amount

to qaT = sn/(si + s2) = 0.85, or 85 per cent of the total. The fre-

quency of ST chromosomes observed in the experimental populations

a year after the start of the selection process is slightly below the

predicted frequency.

Individuals and Populations as Units of Natural Selection. The

experiment just described raises an interesting problem. The outcome

of natural selection in a population containing ST and CH chromo-

somes has been to establish an equilibrium condition at which the

population consists of certain proportions of the chromosomal heterozy-

gotes, ST/CH, as well as of the homozygotes, ST/ST and CH/CH.

These homozygotes, particularly CH/CH, possess a low fitness. We

may say that the CH/CH homozygotes are afflicted with a grave

hereditary disease; a fitness of 41 per cent means that the genotype is

effectively semilethal. Natural selection, then, causes the appearance

in every generation of a certain number of genetically crippled indi-

viduals. It seems very strange that natural selection should fail to

eliminate from the population the unfit genotypes.

The solution of this paradox, however, is simple. Let us keep in

mind that the fittest genotypes in many populations are heterozygotes

endowed with hybrid vigor. But a sexual population consisting en-

tirely of heterozygotes for a pair of allelic genes or chromosomal struc-

tures would produce, according to Mendel's first law, a progeny con-

sisting of only 50 per cent of heterozygotes and 50 per cent of homo-

zygotes. What natural selection does is to establish proportions of the

genotypes at which the average fitness of an individual in the popula-

tion is the highest attainable one, but the high fitness of the popula-

tion as a whole is purchased at the price of producing some genetically

unfit individuals (the homozygotes). We are compelled to conclude

that, with sexual reproduction, it is the Mendelian population, as well

as the individual which is the unit of natural selection and evolution.

148

Individuals, Populations, and Races

The hybrid vigor is a very widespread phenomenon in sexually re-

producing populations of both wild and domesticated animals and

plants. We shall discugs it in more detail in Chapter 9. The fitness

of these populations is maintained by natural selection by means of

mechanisms such as are described for Drosophila populations.

Chromosomal Races in Drosophila. A population of Drosophila

living in nature, like the experimental populations considered above,

_

1

m

W

g^l

r1!

7

10,000

J

\

A

/

f^l

— \

t/x3

8,000

11

^

•

I1

^

/

••*

$

'

7- .-

\

6,000 .E

':::' ::':-: f.::'^i

^

J

:<-:::':-:-:'::

«•»••''"*

***^

A fYWI "£

.-.

J

I

1

jf

Q

*

_^^

y^

o nnn

_J^-

— • —

— 1

1

~1

— i —

0

10

0

30

40

0

50

6

Distance in miles

Figure 7.5. The relative frequencies (symbolized by the dimensions of the

Blood Groups in Man 149

localities in the Sierra Nevada not far removed from Mather, but lying

at different elevations. The sizes of the squares symbolize the fre-

quencies of the different chromosomes; stippled squares represent

frequencies of the chromosomes with the ST gene arrangement; white

squares show the frequencies of AR chromosomes; and cross-hatched

squares refer to the frequencies of CH chromosomes. Now, at the

locality lying at the lowest elevation above the sea level (on the left

in Figure 7.5) the ST chromosomes are the commonest, followed by

AR, and CH chromosomes are relatively rare. As one ascends the

mountain range (moving to the right in Figure 7.5), the ST chromo-

somes become progressively less frequent, and AR chromosomes more

and more frequent.

The situation observed in the lowly Drosophila fly is very instructive,

for it illustrates the nature of races in many sexually reproducing or-

ganisms, including man. Consider that the population of the highest

locality (elevation 10,000 feet above the sea level, the rightmost popu-

lation in Figure 7.5) is very different in composition from the popu-

lation of the lowest locality. But Figure 7.5 shows that localities inter-

mediate in elevation have populations of intermediate composition.

The frequencies of genes and chromosomal variants in the gene pools

of adjacent populations change gradually; they form geographic gradi-

ents or dines.

Geographic Distribution of the Blood Groups in Man. The blood

groups in man reveal a situation remarkably parallel to that shown by

the chromosomal types in Drosophila. As we know, the three alleles

of a gene, 1°, IA, and P, give four blood groups: O, A, B, and AB.

With the exception of some American Indian tribes which are (or

were before the invasion of the Europeans) uniform for blood group

O, most other human populations contain individuals of all four blood

types. The relative frequencies of the four types, however, are by no

means uniform, as shown in Figures 7.6 and 7.7. In the Old World,

the gene /" is most common in Central Asia and in India, and shows

descending gradients both westward and eastward. The gene ZA is

common in Western Europe, in some parts of Africa, in Australia, and

the Blackfeet and Blood tribes of the American Indians (in Montana

and the adjacent part of Canada). The gene 7° is common every-

where, especially among the American Indians.

It is clear that knowing the blood group of an individual human

being (or the chromosomal constitution of an individual Drosophila

pseudoobscura) is, with some exceptions, quite insufficient to deter-

mine his "race" or the geographic location of the population from

152 Individuals, Populations, and Races

which the individual came. For this purpose we must have a sample

of individuals from a population, and the greater the sample, the more

exactly can its geographic origin be inferred.

A question may perhaps be asked at this point. Why not divide the

human species into four races corresponding to the four blood groups,

or the species Drosophila pseudoobscura into a number of races cor-

responding to the several chromosomal types? The answer is simple:

brothers and sisters often belong to blood groups (or chromosomal

types) different from each other and from their parents. For example,

if the two parents belong to O and AB blood groups, respectively, half

of their children will belong to A and the other half to B blood groups.

Furthermore, mankind varies not only with respect to the gene Z°-

7A-7B but with respect to many other genes. Thus some people have

Rhesus positive and others Rhesus negative blood, and this trait is

determined by alleles of another gene, Rh. A person of any of the

four blood groups, O, A, B, and AB, may be either Rhesus positive or

Rhesus negative. Human populations are differentiated with respect

to the Rhesus gene, the Rhesus negative allele being most frequent in

the populations of Western Europe, particularly among the Basques of

northern Spain and southern France. To these must be added several

other known genes which determine the properties of human bloods,

the various combinations of which give several thousand blood geno-

types, each of which would have to be considered a separate race.

But there are also many other genes which are variable in human popu-

lations, the combinations of which are so numerous that no two indi-

viduals are likely to have the same genotype. We come, then, to the

conclusion already stated above: races are not distinct individuals or

genotypes, but populations which differ in the incidence of some genes

or chromosome structures in their gene pools.

The Problem of "Neutral" Race Differences. Human races charac-

terized by blood group frequencies differ in an important respect from

the chromosomal races in Drosophila discussed above. In Drosophila

some of the chromosomal types are adaptively superior to others in

certain environments, and the chromosomal races are adapted to the

environments of the territories in which they occur. But in man there

is little wholly convincing evidence that the differences between hu-

man races are or are not adaptive. Is the straight thick hair charac-

teristic of the yellow race\best suited for living in Asia, the thinner

straight, wavy, or curly hajr of the white race for living in Europe,

and the kinky Negro hair for living in Africa? Is there any advantage

in having the thick Negro lips in Africa, and thinner lips in Europe

The Problem of "NeutraF' Race Differences 153

and Asia? Why should round (brachycephalic) heads be common

among the inhabitants of Central Europe, and long (dolichocephalic)

heads among the northern ai d the southern Europeans? The dark

skin color of many human p

white race develop the protec

pulations appears to be advantageous

for living in hot climates, and the light color in temperate climates.

And yet, although sunburns j re health hazards, most persons of the

ive suntan easily enough and can with-

stand exposure to the tropical sun. There is a possibility that pale

skin colors are advantageous :when sunshine is scarce, because they

facilitate the formation in the organism of the protective antirachitic

vitamin D. The evidence for this is, however, not decisive.

Many races of animal and plant species differ from each other in

colors, in proportions of some \body parts, and in other traits which

do not seem to facilitate the survival or reproduction of their carriers

in the environments in which they occur. Some zoologists and bot-

anists accordingly believe that many, or even most, race differences

are adaptively neutral—they neither help nor handicap their possessors.

There even was an opinion which for some years was current among

both biologists and anthropologists, that the traits used to distinguish

and to classify races and species should be adaptively neutral. Adap-

tively significant differences are, so the argument ran, too easily modi-

fied by the environment, and consequently not reliable as basis of

classification. But if the differences between races and species are

adaptively neutral, how could they be established by natural selec-

tion, which perpetuates useful and eliminates harmful traits? Those

who believe that organic evolution is brought about chiefly through

the action of mutation, gene recombination, and natural selection are

inclined to think that most race and species differences are adaptive,

or at least were so at the time when these races and species were in

the process of formation.

The plain truth is that little research has been done to study the

-possible effects on fitness of the genetic differences among individuals

within human populations or between the populations (races). All

too often it has been assumed that all "normal" human beings function

very nearly similarly. This typological thinking has doubtless pre-

vented the discovery of the adaptive significance of numerous human

traits. It should be remembered that even slight selective advantages

may be important in evolution. If the possessors of darker skins pro-

duce, in the tropical climates, an average of 101 surviving children for

every 100 children produced by persons with lighter skins, the tropical

countries will eventually be populated by dark races.

154 Individuals, Populations, and Races

Furthermore, selective advantages need not necessarily be expressed

in robust health. Suppose that in certain cultures the possessors

of curly hair will be sexually more attractive to the opposite sex than

individuals with straight hair. This preference might lead to the

formation of a race with curly hair.

Finally, the externally visible racial "traits" may be only outward

"marks" of physiological differences which are important in adapta-

tion. As pointed out in Chapter 2, a gene difference often produces a

whole complex of traits, some more and others less striking to an ob-

server's eye. An excellent example of such a manifold effect of genes

has been discovered in the common garden onion by Walker and his

collaborators (see Chapter 5). The resistance or susceptibility of the

onion bulbs to smudge (a parasitic fungus) is determined by the pres-

ence or absence in the onion scales of certain chemical substances

(catechol and protocatechuic acid). White onions do not contain

these substances and are susceptible to the smudge; colored (red or

purple) onions are resistant. The color of the onion bulb is known

to be determined by several genes. Is, then, the color of the onion

bulb adaptively important? It may well be that being white or pur-

ple is of no importance to the welfare of the plant. However, the

"color genes" also cause resistance to smudge infections, and this re-

sistance may be very important for the plants growing in localities in

which these infections occur. It is most probable that many allegedly

"neutral" genes in man will be found to induce physiological charac-

teristics of importance in health or disease.

Experiments on Mountain and Valley Races of Plants. Although

we are often unable to see the selective advantages of racial traits,

there is no doubt that the process of formation of races in general

serves to adapt different populations of a species to live in the environ-

ments of different countries in which these populations occur. Ex-

periments bearing on this problem have been made on races of certain

California plants by Clausen, Keck, and Hiesey.

Figure 7.8 shows races of yarrow (Achillea) from different parts of

California. The race growing on the humid Pacific Coast near San

Francisco is a compact plant, with a thick stem, which grows through-

out the year, including the mild winters in the maritime climate of this

race. In the hot and dry interior valley of California there is a race

consisting of very tall plants with gray-pubescent leaves, which grow

chiefly during the winter and spring seasons when there is some rain,

and are dormant during the long rainless summer. In the forest belt

of the Sierra Nevada grows a relatively tall race with a slender erect

Mountain and Valley Races 155

stem. Because the winters in its native habitat are severe the plants

become dormant when the weather turns cold, and the active growth

and flowering occur during the spring and summer. Finally, in the

high alpine zone of the Sierra Nevada is a dwarf race which is able

to grow, flower, and mature its seeds during the very short mountain

summer, and to remain dormant over the long cold season.

The question which naturally arises is this: To what extent are the

differences between the races of yarrow due to diversity of their geno-

types, and to what degree are they induced directly by the environ-

ments which these races inhabit? With yarrows, this question can be

resolved by experiments more conclusively than is possible with hu-

man races. A yarrow plant can be cut in two or several parts; the

parts may be replanted in soil, where they continue to grow and

eventually flower. The parts have the same genotype; they are mem-

bers of a clone. The differences which may appear among the plants

growing from parts of the same individual replanted in different lo-

calities are due to the environments of these localities. In the experi-

ments of Clausen, Keck, and Hiesey, individuals of the different races

of yarrow were subdivided, each into three parts, and the parts were

replanted in three experimental gardens. One of the gardens lies in

the coastal region of California; the second, at a moderate elevation

above the sea level in the forest zone of the Sierra Nevada; and the

third high in the alpine zone of these mountains.

The coastal race of the yarrow planted in the forest zone of the

mountains is forced into dormancy during the winters, grows slowly

during the spring and the early summer, and flowers much later than

the native race does. The coastal race is usually killed in the alpine

zone during the first winter, and it never develops flowers; the geno-

type of the coastal race may be said to be lethal in the alpine zone.

The race native in the forest zone of the mountains grows quite tall

when planted in the coastal garden, but about two-thirds of the indi-

viduals become dormant during winter. In the alpine garden this race

shows a variable behavior: some individuals die, others grow only

slowly, and still others grow tall and come into flower, although too

late in the season to ripen seed. Finally, the alpine race can grow

and flower at both the coastal and the mountain gardens, but it is the

only race which is able to cope successfully with the rigors of its

native habitat. On the coast and in the forest zone the alpine race is,

however, much weaker than indigenous plants.

There is no doubt, then, that the yarrow races have different heredi-

tary endowments. On the other hand, both the external appearance

158 Individuals, Populations, and Races

and the physiological functions of individuals of each race are pro-

foundly modified by transplantation to foreign environments, as Figure

7.8 clearly shows. It is neither heredity alone, nor the environment

alone, which makes a yarrow race what it is. It is the developmental

pattern engendered by a certain heredity in a certain environment.

Some Rules of Racial Variation. If we examine races of several

more or less related species inhabiting the same country, some re-

semblances between them are often noticed. For example, many-

species of beetles are represented in Arizona and in California by races

which are colored more lightly than the races of the same species in

the humid Pacific Northwest and in western Canada. It is as though

the inhabitants of a given country followed a common "style." The

existence of such styles or trends was noticed by zoologists and bot-

anists of the past century. They interpreted these observations in a

way which is not acceptable at the present time: they assumed that

the geographic environment directly alters in some mysterious way

the genotypes of diverse organisms. The problem has been re-exam-

ined more recently by Rensch, Mayr, and others. These investigators

concluded that some, and probably all, rules of racial variation result

from parallel development by natural selection of analogous adapta-

tions in different species of organisms.

It is easy to see that natural selection in cold climates will favor in

diverse animals adaptations which minimize the heat loss; in hot cli-

mates arrangements will be favored which facilitate the cooling of the

body. Accordingly, the races of many mammals that live in cold

countries have longer but finer fur than races of the same species in

hot countries. Thus for the mountain lion (puma) in the relatively

cool climate of the mountains of Mexico the mean length of the hair

in the pelage is about 31 millimeters, whereas for the race of the

same species in the equatorial climate of the Amazon valley the mean

hair length in the comparable part of the pelage is only 12 millimeters.

The European fox has fur 46 millimeters long in eastern Germany, and

in subtropical Algeria only 39 millimeters. The otter fur in eastern

Germany is 23 millimeters long, and in tropical Ceylon only 15 milli-

meters long.

Races of many animals (warm-blooded as well as cold-blooded

ones) which inhabit cold climates tend to be larger in body size than

races which live in warm countries. Moreover, races of warm-blooded

animals in cold countries have relatively shorter tails, legs, ears, and

beaks than the inhabitants of warmer climates. The biological sig-

nificance of these rules is understandable if we consider that a larger

Major and Minor Races 159

body has a relatively smaller body surface than a smaller body of the

same shape. The body surface loses heat by radiation and by convec-

tion; accordingly, a relatively small body surface will be favored in

cold climates, and a large body surface in hot climates. We need only

to recall the discomfort felt in our ears in very cold weather to under-

stand that relative reductions in size of such protruding body parts

as ears, legs, and tails will be favored by natural selection in cold

climates. In hot climates, on the contrary, these protruding body

parts will facilitate the cooling of the body.

Major and Minor Races. Authorities in anthropology often disagree

on just how many races compose mankind. Two to more than two

hundred races have been distinguished by different specialists in race

studies. So wide a divergence of opinion arose because race differ-

ences are quantitative rather than qualitative. Some races differ very

slightly, others are quite distinct (see Chapter 13).

We have seen that the populations of Drosophila pseudoobscura

which live at different elevations in the Sierra Nevada of California

differ in the incidence of certain chromosomal structures (Figure 7.5).

The same kinds of chromosomal structures occur at all elevations, al-

though in different proportions. Much greater differences have been

found between populations of this species of fly in California and in

central Mexico. Most chromosomal structures which occur in Cali-

fornia do not occur in Mexico, and vice versa; those which do occur

in both regions have different frequencies. It can be stated that popu-

lations which live at different elevations in the California mountains

are racially distinct from each other. Populations of California are

racially distinct from populations of Mexico, but the differences be-

tween the California populations are much smaller than those be-

tween California and Mexico populations.

In man, the native populations of Europe (white), of central Africa

(Negro), and of central Asia (Mongoloids) are obviously different,

even to a superficial observer. The populations of different European

countries are also different, but much less strikingly so. All these

populations are racially distinct. Should we conclude that there exist

"major" and "minor" races? In a way this is a correct description of

the situation, but we must emphasize that there are also all possible

transitions between very slightly and very strongly marked racial dis-

tinctions. Race differences of all magnitudes may be encountered.

To a classifier of races this fact creates great difficulties, since it makes

agreement on the number of races to be recognized virtually unattain-

able. To an evolutionist, however, the same fact is most significant.

160 Individuals, Populations, and Races

It shows that race formation is a gradual process, which leads to a

progressive divergence of populations in response to the environmen-

tal differences encountered in the territories which those populations

inhabit.

Are Races "Real" or "Man-Made"? The abuse of the race concept

by politicians has made many people skeptical of the usefulness of

distinguishing any races at all. Indeed, do races actually exist as bio-

logical entities? Or are races arbitrary subdivisions which anthro-

pologists and biologists have made for their own convenience? A

proposal has also been made to abolish the term "race" altogether,

and to distinguish instead "ethnic groups" in the human species. The

argument in fr.vor of this course is that the word "race" has become

charged with too much emotion and prejudice.

It may well be doubted whether race prejudice can be effectively

combated by calling races ethnic groups or by some other names. A

better way wculd seem to be to explain to people the elementar)

biological facts underlying the scientific race concept. It is a plain

fact, obvious to any reasonable observer, that mankind consists of

populations, chiefly allopatric ones, which differ in the incidence of

various genetic characters. The genetic differences between human

populations are of the same kind as can be observed between races

or subspecies of domestic and wild species of animals and plants. -

How Many Races? Much of the misunderstanding surrounding the

concept of race is due to people's confusing two logically and methodo-

logically separate problems. The first problem i$ a biological one, and

it may be presented as a question: "Are the populations which inhabit

certain countries genetically, and hence racially, distinct?" The second

problem is a question of classification and nomenclature: "Do certain

populations differ sufficiently, to deserve being given different names?"

The first problem can, with adequate study, be solved quite objectively.

If the populations in question do differ in the incidence of some

genetic variants in their gene pools, they are racially distinct. Race

differences are manifestly real and ascertainable' facts. But the dif-

ferences between the fly populations of adjacent sections of a forest,

or between human populations of neighboring towns or districts, are

very small; the differences between remote populations are large. It

would be most inconvenient to give a separate racial name to every

local population, and would merely burden the scientific literature.

It is, then, up to the investigator to decide how great should the racial

differences be to justify giving them names. It is arbitrary how many

races we distinguish in a species, or rather it is a matter of con-

How Many Races?

161

venience and common sense. For some purposes, for example for an

elementary course of anthropology, it is better to recognize only a small

number of "major" races. For a detailed study of the inhabitants of

a country a finer subdivision may be called for. The human species,

and other species, contain as many races as we see fit to distinguish

(see Chapter 13).

Figure 7.9. Races of the golden whistler (Pachycephala pectoralis) which live

on different islands of the Solomon Archipelago, in tropical Pacific. (After Mayr.)

However, there is a biological phenomenon which may be made use

of in order to make the racial groups more objective than they would

otherwise be. One of the major difficulties with race classifications

has been that, although the geographically remote populations are

clearly distinct, the intermediate populations connect the extremes by

insensible gradations. As we have seen, these gradations are due to

the existence of gene gradients or clines. It frequently happens, how-

ever, that the gradients are not even; they do not consist of so

much percentage increase, or decrease, of gene frequencies per every

mile, or every hundred miles, traveled. Instead, in some geographic

regions the gradients are more abrupt, in others they are less steep.

For example, the human populations which reside on the two sides of

162 Individuals, Populations, and Races

the Himalaya Mountains or those living north and south of the Sahara

Desert are fairly distinct. The reasons for this are not far to seek.

Not only is the environment likely to be rather different on the two

sides of a major geographic barrier, but such a barrier impedes the

travel and migration between the populations which it separates. The

racial divergence, then, is accelerated by geographic barriers. The

race classifier turns this situation to his advantage by making his di-

viding lines between races or subspecies coincide with the zones of

the steep gene gradients. The Himalayas separate the Hindu branch

of the white race (or the Hindu race) from the yellow race; the Sahara

separates the white and the Negro races.

Genetic Differences and Adaptation. To some people uniformity

has a considerable emotional appeal, especially where man is involved.

To gratify this emotion, J. J. Rousseau (1712-1778) invented his theory

of "tabula rasa," according to which a newborn infant is a "blank

page" on which environment and education write this or that story of

the individual's life. Somewhat similar views have more recently been

espoused by the behaviorist school of psychologists, as well as by

certain fashionable exponents of psychoanalysis and cultural anthro-

pology.

In part, the appeal of all these variants of the "tabula rasa" theory

is due to the confusion of biological uniformity with social, legal, and

religious equality. The idea of equality is certainly precious, since it

is the basis of democracy and of humanitarian thinking in general. But

this idea is derived not from biology but from the Christian tradition,

which is the basis of Western civilization. People are most certainly

not biologically alike, but they need not be so in order to be equal

before the Law and before God. In fact, democracy may be regarded

as an arrangement which permits unlike and yet equal persons to live

together and to collaborate for the good of society and of mankind

(see Chapter 14).

It is an ascertainable fact that the human species, as well as probably

all other sexually reproducing species, show a greater or lesser amount

of genetic diversity. This diversity is expressed within a Mendelian

population in genetic differences among individual members of a breed-

ing community. It is expressed in space by formation of allopatric or

geographic races or subspecies, and it is expressed in time in changes

which Mendelian populations undergo from generation to generation

in the process of evolution.

There can be no doubt that most of the organic diversity in the

Suggestions for Further Reading 163

sympatric, allopatric, and temporal levels has an important biological

function to perform in making the populations adapted to live in dif-

ferent environments. No genotype is perfect, in the sense of making

its carriers ideally adapted to live in all possible environments. A

genotype may be superior to others in a certain environment but in-

ferior in other environments. A genetically uniform species would be

able to exploit successfully only a few ecological niches available in

the territory where it lives. A polymorphic species can occupy a

greater number of habitats. A species differentiated into races, other

things being equal, may occupy a greater territory than a single race

could. And a species which alters its genetic constitution when the

environment changes, again other things being equal, is more likely to

endure than a rigidly fixed species. It is most probable that, on the

human level, the polymorphism within, and the race differences be-

tween, the populations have furnished the creative leaven for the cul-

tural development of mankind.

Suggestions for Further Reading

MayV; E. 1942. Systematics and the Origin of Species. Columbia University

Press, New York.

This work has played an important role in achieving a synthesis of zoological

systematics with the modern theory of evolution, which was originally based

chiefly on the findings of genetics. A parallel, but more inclusive, work dealing

with botanical systematics is Stebbins' book, quoted among the suggested readings

in Chapter 6. The books by Andrewartha and Birch and by Dobzhansky, also

quoted in Chapter 6, may also be useful for understanding the race and species

concepts as used in modern biology.

Clausen, J. 1951. Stages in the Evolution of Plant Species. Cornell University

Pre^sflthaca.

fcJiKisen, J., Keck, D. D., and Hiesey, W. M. 1948. Experimental studies on the

nature of species. III. Environmental responses of climatic races of Achillea.

Carnegie Institution of Washington Publication 581.

Dobzhansky, Th., and Epling, C. 1944. Contributions to the genetics, taxonomy

and ecology of Drosophila pseudoobscura and its relatives. Carnegie Institution

of Washington Publication 554.

Lack, D. L. 1947. Darwin's Finches. Cambridge University Press, London.

Patterson, J. T., and Stone, W. S. 1952. Evolution of the Genus Drosophila.

M;ii-ini!l.M' New York.

These five books (Clausen to Patterson and Stone) describe experimental and

observational studies on races and species of organisms which are particularly

favorable for such studies.

164 Individuals, Populations, and Races

Count, E. W. (Editor). 1950. This Is Race. Schumann, New York.

This is an anthology of anthropological writings concerned with the race con-

cept as applied to man. With some exceptions, the/writings included belong to

the pregenetical period.

Boyd, W. C. 1950. Genetics and the Races of Man. Little, Brown, Boston.

Application of the genetical concept of race to man.

on, C. S., Garn, S. M., and Birdsell, J. B. 1950. Races: a Study of the Prob-

\J lems of Race Formation in Man. Ch. Thomas, Springfield, 111.

A classification of human races and a discussion of the possible adaptive sig-

nificance of human racial characteristics.

v Origin and evolution of man. 1950. Cold Spring Harbor Symposia on Quantita-

l*/ five Biology, Volume 15.

The volume contains articles by thirty-seven authors, dealing with the prob-

lems of human polymorphism, race formation, and other aspects of human evo-

lution.

8

Species

Darwin's great book was entitled The Origin of Species (the full

title was rather more ponderous: On the Origin of Species by Means

of Natural Selection, or the Preservation of Favoured Races in the

Struggle for Life). Darwin knew as well as anybody that formation

of races and species is only a part of the grand story of evolution. The

origin of species, however, was of crucial importance in Darwin's day,

because of the current view that species were created entities which

could not be produced by natural processes.

In a different way, species formation is also regarded as a critical

stage of the evolutionary process in modern theories. Races of a spe-

cies are populations which can, and often do, cross and exchange genes.

Hybridization of races may, and sometimes does, lead to their fusion

in a single population. This has actually happened, for example, to

some human races, the members of which intermarried so frequently

that the races disappeared as separate Mendelian populations. Races

are genetically open systems, and the divergence of races is a re-

versible process; it can be undone by hybridization. Species, on the

other hand, are genetically closed systems, since they exchange genes

rarely, or not at all. Evolutionary divergence of Mendelian popula-

tions tends to become irreversible once the species level is reached.

For example, man and chimpanzee are most unlikely ever to exchange

genes or to form a hybrid population.

A minority of modern evolutionists, among whom Coldschmidt is

most prominent, believe that the known factors of evolution—mutation,

gene recombination, selection, and genetic drift—account only for

"microevolution," which is usually enuated with race formation. Other,

165

166 Species

and as yet unknown, processes should explain "macroevolution"—the

origin of species and species groups.

Hypothesis of Constancy of Species. Living beings seem infinitely

variable; no two individuals are completely alike. But the variation

is not entirely haphazard, since on closer acquaintance living creatures

are seen to fall into discrete groups. Everyday language recognizes

the existence of these discrete groups, refers to them as "kinds" of ani-

mals or of plants, and gives to each "kind" a vernacular name—cat, lion,

tiger, jaguar, and puma (mountain lion), etc. Linnaeus (1707-1778),

the father of systematic biology, called the "kinds" species, and gave

to each species known to him a name in Latin. The species referred

to above are Felis maniculata, Fclis leo, Felis tigris, Felis onca, and

Felis concolor, respectively.

The belief in constancy of species was not accepted by everybody

before Darwin. Weirdest transformations of species were often cred-

ited. Greek and other mythologies are full of stories of gods trans-

forming themselves or men into animals and back into men or gods.

Theophrastos in ancient Greece, Pliny in ancient Rome, Albertus Mag-

nus in the thirteenth century, Paracelsus in the sixteenth, Telliamed

in the seventeenth, and Lysenko in the twentieth, all believed in trans-

formation of one species into another. They thought that wheat occa-

sionally produces rye or barley, that fishes turn into birds, etc. Oppian

(third century Rome) reached the peak of absurdity: he believed

that the ostrich arose from a cross between a camel and a sparrow.

Linnaeus faced the enormous task of classifying the organisms and

making the immense variety of living forms intelligible. To him, the

species was the elementary unit of classification, and supposing that

this unit was fixed and unchangeable was a tenable opinion. He stated

this opinion in his famous dictum: "Species tot sunt, quot diversas

formas ab initio produxit Infinitum Ens" (Species are as many as were

produced at the beginning by The Infinite). It is important to real-

ize that in Linnaeus's day such a view did not appear contrary to facts,

as it so clearly does now.

Development of the Species Concept from Linnaeus to Darwin. In

1758 Linnaeus knew 4235 species of animals. At present approximately

1,000,000 animal and about 265,000 plant species have been described.

Furthermore, Linnaeus worked mostly in Sweden and in other coun-

tries of northwestern Europe, and his collections came naturally from

that part of the world, with only a scattering of specimens from other

countries. In other words, Linnaeus worked chiefly with sympatric

(page 137) species. It happens that, for reasons explained below,

Development of the Species Concept 167

sympatric species are usually quite discrete, no individuals intermedi-

ate between the species being found. If we collect animals or plants,

for example, in the vicinity of New York, we usually find little diffi-

culty in classifying the specimens into clear-cut species. The gaps

between the species appear to be absolute and unbridgeable. To such

species the hypothesis of constancy seems to apply reasonably well.

The late eighteenth and the nineteenth centuries were the times of

rapid geographical exploration of the world. Collections of animals

and plants were made in near and remote lands, and were deposited

in the museums of Europe and of America. Zoologists and botanists

had more and more allopatric (page 137) forms of life to classify.

Here difficulties began to arise and to multiply, until the successors

of Linnaeus saw themselves forced to abandon the idea that species

are fixed.

We have seen, in Chapter 7, that populations of the same species

which live in neighboring territories usually differ only slightly in the

incidence of some genes and of bodily traits which these genes de-

termine. Races found in remote territories, or in territories which are

separated by barriers making migration difficult, may be more sharply

distinct. If we knew only the native inhabitants of Europe and of

central Africa, they would appear to us very different. But if we

study people in all parts of the world, we find all kinds of races

intermediate between those of Europe and central Africa. Remote

races of many species differ even more than remote human races. In

fact, remote races of a species may appear about as distinct as

separate species.

Lamarck, whose long life was spent in classifying species of plants

and animals, was perhaps the first to see clearly that a new working

hypothesis was demanded by the new evidence. This hypothesis was

that species evolve from races. Races and species are stages in the

evolutionary divergence, produced by the adaptation of the organisms

to their environments. Lamarck's contemporaries were not convinced

that so radical a hypothesis was necessary, but some thirty years later

Darwin came to the same conclusion: "In all these several respects the

species of large genera present a strong analogy with varieties. And

we can clearly understand these analogies if species have once existed

as varieties, and have thus originated; whereas, these analogies are

utterly inexplicable if each species has been independently created."

It is at this point immaterial that Lamarck and Darwin had different

views concerning the processes which bring about the formation of

new races and species. Lamarck thought that new races and species

168 Species

were formed chiefly through inheritance of the results of use and dis-

use of organs, whereas Darwin considered natural selection to be the

most effective agent. They agreed that the absence of any dividing

lines between races and species shows that races are incipient spe-

cies. Thus the difficulty of making clear-cut classifications, which

systematists found so annoying in their work, became the foundation

of probably the greatest discovery which biological science has yet

made.

Theory of Geographic Speciation. Although Darwin was quite

familiar with species composed of geographic races, it remained for

his successors to develop a theory of geographic species formation

(speciation). This theory was stated by M. Wagner (1868 and later),

Jordan (1905), and in our day by Rensch (1929) and by Mayr (1942).

Any species tends to spread over the surface of the earth into all

territories where it can live. Populations of a species may thus be

exposed to different climatic or soil conditions, to different predators,

parasites, and diseases, and generally to varying environments.

Through natural selection, populations respond adaptively to the

environmental differences, by becoming genotypically and pheno-

typically differentiated into races or subspecies. It stands to reason

that geographically remote territories are likely to have more sharply

different environments than adjacent territories. Accordingly, remote

territories are likely to be inhabited by most distinct races. Some of

the allopatric races may gradually become very different in genetic

structure and in external appearance from other races. If and when

this happens, the single ancestral species breaks up into two or sev-

eral derived species. When the new species are fully formed, they

may invade each other's territories, come to live side by side, and thus

become sympatric species.

Most biologists are now agreed that the usual method by which one

species may split up into two or more species is by way of divergence

of geographic races. Whether this is the only method is questionable.

Some investigators believe that populations may differentiate and

become species by becoming specialized to eat different foods or to

grow on different soils in the same territory, sympatrically. It is,

indeed, probable that such sympatric species formation occurs in

organisms which are able to reproduce asexually or by self-fertiliza-

tion. Species formation by allopolyploidy (Chapter 9) is necessarily

sympatric.

Geographic and Reproductive Isolation. Different races of man

and different breeds of domesticated animals and plants often live

Geographic and Reproductive Isolation 169

sympatrically, in the same territory. Biologically, this is an unusual

situation, since races of sexual and cross-breeding organisms are nor-

mally allopatric, but it is easy to see how it happens. In man the

biological urge to mate is channeled by social custom. The regula-

tion of marriage by custom has the biological result of slowing down

the exchange of genes between different mating groups, such as social

or economic classes and sympatric or allopatric races. In domesticated

animals and plants the control of reproduction is exercised by man.

Consider the races (breeds) of dogs, many of which live sympatrically.

The gene exchange between them is restricted or prevented by man.

When this restriction is weakened, the once separate sympatric races

fuse into a single variable race—the mongrel.

In the state of nature the gene exchange between races of sexual

organisms is limited by distance, by separation in space, by geographic

isolation. This is why most races are allopatric. Neighboring allo-

patric populations usually interbreed more or less frequently, but the

diffusion of genes from one race to other remote races is slow enough,

so that the races remain distinct. Migration or diffusion of genes tends

to make populations or races converge; natural selection and genetic

drift tend to make them diverge. Actual convergence or divergence

depends upon which of these factors get the upper hand.

Different, even closely related, species are often sympatric. For

example, Patterson and Stone found that about forty species of Dro-

sophila live together in one locality near Austin, Texas. Nevertheless,

hybrids between these species are rare or absent. Why do races

hybridize and exchange genes wherever possible, whereas species do

not? The answer is that gene exchange between species is prevented

by a variety of causes, known collectively as reproductive isolating

mechanisms, in addition to geographic isolation which may or may

not be present.

Distinct species may live either in the same or in different terri-

tories; both sympatric and allopatric species are common. The ex-

change of genes between sympatric species is limited or prevented by

reproductive isolation; that between allopatric species, by geographic

as well as by reproductive isolating mechanisms. For example, the

wild ancestors of the domestic horse and of the domestic ass may have

been in part sympatric. Under domestication, hybrids between them,

mules, can be obtained, but mules are almost invariably sterile. Gene

exchange between the populations of horses and asses is absent. The

species are reproductively isolated. On the other hand, wild horses

and zebras were allopatric species, since the former occurred, in his-

Two Species of Drosophila

171

obscura is more frequent in warmer and drier locations, and at lower

elevations in the mountains, than D. persimilis. The habitat isolation,

however, is incomplete, and in many places the two species occur side

by side. Their hybridization is impeded, however, by sexual isolation,

that is, by preference for mating within, rather than between, the

species, as can be demonstrated by a simple experiment. A mixture

of virgin females of both species is exposed to males of one of them.

The males inseminate a greater proportion of females of their own

than of the foreign species (Table 8.1). If females of either species

TABLE 8.1

SEXUAL ISOLATION BETWEEN Drosophila pseudoobscura AND Drosophila persimilis

(Equal numbers of females of the two species were exposed to males of one of

them. The numbers show the percentages of the females inseminated and left

virgin in such experiments.)

\. Females

Males \.

pseudoobscura

persimilis

Inseminated

Virgin

Inseminated

Virgin

pseudoobscura

persimilis

84.3

22.5

15.7

7.0

79.2

93.0

20 8

77.5

are offered a "choice" of mating with males of both species, they mate

with conspecific males much more frequently than with foreign males.

Observations in nature have shown that under natural conditions

the sexual isolation between the species is even stronger than the

experimental data in Table 8.1 might suggest. Females inseminated

by males of foreign species are exceedingly rare. This fact may be

due in part to an ethological isolation, the two species being somewhat

different in behavior: D. pseudoobscura is relatively more active dur-

ing the afternoon and D. persimilis during the morning hours. Yet, the

ethological isolation (of which sexual isolation is a part) is also in-

complete, since some individuals of both species fly both in the morn-

ing and in the afternoon.

Hybrids between D. pseudoobscura and D. persimilis can be rather

easily obtained in laboratory cultures, if females of one species are

confined with males of the other. The hybrid flies of both sexes are as

172 Species

vigorous as the pure species, just as mules are not inferior in vigor

either to horses or to donkeys. Again, like mules, the hybrid male flies

show a complete hybrid sterility. They fail to produce any functional

spermatozoa, on account of gross disturbances in their reproductive

cells (see page 206). The female hybrids, on the contrary, are fully

fertile; they deposit numerous eggs, which, when fertilized by males

of either parental species, may produce viable larvae. However, the

backcross hybrid progenies so obtained suffer a hybrid breakdown;

they are so deficient in vigor and vitality that they survive only under

favorable conditions in laboratory cultures.

To summarize: The gene exchange between D. pseudoobscura and

D. persimilis is impeded by cooperation of at least five reproductive

isolating mechanisms. None of these mechanisms is by itself sufficient

to prevent completely all the gene exchange between the species; yet

the combination of them does accomplish this function in nature. In

laboratory experiments some gene exchange can be obtained, however.

It is not an uncommon situation that species which never cross in

nature can sometimes be crossed in experiments.

Reproductive Isolation between Species of Animals. Detailed stud-

ies of the reproductive isolation between various species of Drosophila

have been made, especially by Patterson, Stone, and Spieth. They

have revealed two significant facts. First, the gene exchange between

species is apparently always prevented not by one but by a combina-

tion of several isolating mechanisms. Second, different pairs of spe-

cies are kept apart by different combinations of isolating mechanisms.

By and large, sexual isolation is most widespread. Spieth found that

almost every species which he studied has its own characteristic

methods of courtship and copulation. And yet, some undoubtedly

distinct species, especially allopatric ones which never meet each other

in the state of nature, show little or no sexual isolation. As a rule,

males are less discriminating than females: whereas males may court

females of their own and of a foreign species, females accept males

of their own species more easily than they do other males.

Some pairs of species show gametic isolation, which seems to be

absent between Drosophila pseudoobscura and D. persimilis, discussed

above. The sperm delivered by a male into the genital ducts of a

female is stored in a special organ, the seminal receptacle. When the

male and the female belong to the same species, the sperm in the semi-

nal receptacle remains alive for days and even weeks. But when a

male of Drosophila virilis impregnates a female of D. americana (or

male D. americana, female D. virilis) the sperm loses its fertilizing

Reproductive Isolation between Species of Animals 173

ability a day or two later. Some species of Drosophila have the so-

called insemination reaction: the copulation is followed by a swelling

of certain parts of the genital ducts of the female, which temporarily

prevents the deposition of eggs and new copulations. The insemina-

tion reaction produced by an insemination by the sperm of a foreign

species may be so violent that the females become permanently sterile;

yet other species of Drosophila show no trace of insemination reaction.

A very common isolating mechanism is hybrid inviability. Fertili-

zation of eggs of one species by spermatozoa of another may be accom-

plished, but the hybrid zygote may fail to develop normally. Thus a

belief was long current that domestic goats occasionally produce hy-

brids with domestic sheep. Controlled experiments, with both natural

and artificial insemination, showed, however, that cross-fertilization

between these species does take place, but the hybrid embryos are

aborted at a very early stage of pregnancy. It is virtually certain that a

similar fate overtakes the hybrids between rabbits and hares, although

alleged viable hybrids of this kind have been repeatedly claimed since

1780, and were even given a name—"leporids." In many marine ani-

mals no copulation occurs; eggs and spermatozoa are simply released

into ambient water, and fertilization occurs outside the bodies of the

parents. This external fertilization makes possible experimental hy-

bridization of some very unlike organisms. In the classical experi-

ments of Godlewsky (1906) and Baltzer (1910), several different

genera of sea urchins were intercrossed, and sea urchin eggs were fer-

tilized with the sperm of a sea lily. Sea urchins and sea lilies belong

to different classes of the phylum of echinoderms. Such remote hy-

bridization never yields normally developing hybrid embryos. Either

the chromosomes of the father, brought in by the spermatozoon, are

thrown out of the egg nucleus and disintegrate, in which case the

embryo develops as that of the species of the mother. Or else, if the

parental chrt>mosomes divide and are included in the cleavage nuclei,

the embryos show gross abnormalities and soon die off.

Isolating mechanisms which operate between species of higher ani-

mals are pretty much the same as those found between species of

Drosophila and other invertebrates. As in Drosophila, the precise

situation varies endlessly from species to species. Sexual isolation is

often important, since the courtship behavior is generally unlike even

in closely related species of birds and mammals. This sexual isolation

may account for the fact that species hybrids are on the whole quite

rare in the state of nature, although they are often obtained fairly easily

in experiments. Thus, in captivity, many species and even genera of

174 Species

ducks produce hybrids, some of them quite viable and fertile, and yet

very few hybrid ducks have ever been found in natural habitats. Dice

and Blair found the same to be true for some species of deer mice

(Peromyscus).

Various crosses have been made in captivity between domestic

cattle (Bos taurus), American buffalo (Bos americanus), European

buffalo (Bos bonasus), Asiatic yak (Pheophagus grunmeus), and In-

donesian banteng (Bibos banteng). The hybrids are fully viable,

although difficulties occur at parturition when a relatively small mother

(domestic cow) gives birth to calves sired by large fathers (European

buffalo). The FI hybrid bulls are completely sterile, but the cows are

fertile even when intergeneric hybridization is involved, as in the case

of cattle X yak. The hybrid cows, then, can be backcrossed to either

parental species; among the resulting progenies, the females as well

as some of the males may be fertile. Indeed, some introgression (see

page 127) of genes from the yak into populations of cattle has taken

place in some parts of central Asia, where the two species are kept

together as domestic animals.

Reproductive Isolation in Plants. Among plant species we again

meet a variety of isolating mechanisms. Hybrid inviability and hybrid

sterility occur in plants as well as in animals. An interesting example

of the former is hybrids between certain species of flax described by

Laibach (1925). The hybrid embryos are so weak that they are un-

able to break the outer seed coat, and they die within the seed. Yet

when they are artificially freed from the seed coat they give healthy

seedlings and mature and fertile plants. Blakeslee and Satina (1944)

found an even more extreme situation among hybrids between cer-

tain species of Jimson weed (Datura), where the hybrid embryos

have to be cultivated on artificial media supplied with certain chemical

substances, but may eventually grow normally. Some sterile plant

hybrids will be discussed in the next chapter.

Some plant species are able, in experiments, to produce viable and

fertile hybrids, but do not do so, or do only rarely, in nature. Their

success may depend upon habitat isolation, as for two species of

violets, Viola arvensis and Viola tricolor. The former species grows

chiefly on calcareous soils; the latter occurs mostly on acid soils. The

two species seldom meet (Clausen 1922). In the Yosemite region of

the Sierra Nevada of California, the manzanita bush, Arctostaphylos

mariposa, grows at lower elevations in the mountains than the related

species, Arctostaphylos patula. In some places, however, the alti-

Single Genes and Gene Systems 175

tudinal ranges of the two species overlap, and there some hybrid

bushes can be found.

The isolation of some species is due wholly or in part to the fact

that their flowers attract different animals to transport their pollen.

Grant (1951) found that the columbine Aquilegia formosa in the

mountains of the Sierra Nevada is pollinated chiefly by humming

birds, whereas the related Aquilegia pubescens attracts chiefly hawk

moths. But bumblebees visit both species, and as a consequence

some hybrids are produced in natural habitats. An extreme case of

insect specificity has been discovered among certain orchids in North

Africa. These species have a part of the flower resembling in shape

and in color the females of species of wasps or bees; the male insects

are attracted to these flowers and engage in copulatory movements,

during which they receive and deposit the pollen of the plant. But

different orchids resemble different species of wasps, making hybridi-

zation unlikely.

If flowers of one species receive the pollen of another species, hy-

brids are by no means always produced. In some plants the pollen

of one species grows poorly in the style of a different species, and

fails to reach the ovules in time to accomplish the fertilization. This,

in addition to the inviability of the hybrid seeds, prevents the gene

exchange between some species which grow in close proximity and

are pollinated by unselective agents, such as wind.

Single Genes and Gene Systems. For the sake of simplicity, we

have so far considered only the simplest microevolutionary changes,

which involved substitution of one or of few genes. The transforma-

tion of colon bacteria susceptible to bacteriophage attack into resist-

ant ones requires apparently a single gene change (Chapter 5). The

same is probably true of the resistance of these bacteria to strepto-

mycin, and of the resistance of the red scale to cyanide gas. When,

as in the development of the resistance to penicillin in Staphylococcus

bacteria, the modification occurs gradually, the process is due to

accumulation of several mutations, each of which increases the re-

sistance by a small amount (Chapter 5). Such mutations are said

to be additive.

But evolution often involves more complex situations, some of which

must be considered to gain an understanding of the biological meaning

of species formation. The development of a living body, especially

in higher organisms, involves joint action of all the genes which the

body possesses. In evolution, the gene system as a whole undergoes

changes, although the changes of the gene system are compounded

176 Species

of changes in individual genes. A gene system may be likened to

a mosaic picture, and the genes to the component stones. The nature

and quality of a mosaic picture are determined obviously by the pat-

tern in which the stones are placed, as well as by the characters of

the separate stones.

In Chapter 2 we have considered as an example of interaction of

genes the determination of the coat color in horses. Here many genes

are involved, the color being the result of co-action of all of them.

Particularly interesting is the gene I, which in the presence of the

gene B transforms the black coat color into bay, but in the absence of

B has no known effects. The bi and bl horses are both chestnut

(sorrel) in color. Suppose, for the sake of argument, that having the

bay color is adaptively advantageous to the horse; should the gene 7

then be considered favorable or neutral? Obviously, this would de-

pend upon the other genes which the organism has, particularly on

the gene B.

Another example of gene interaction with which we are familiar is

the chromosomal variants in natural populations of Drosophila pseudo-

obscura (Chapter 7). These chromosomes differ, it may be recalled,

by inversions of blocks of genes in certain chromosomes. The hetero-

zygotes which carry the two chromosomes of a pair with different gene

arrangements are heterotic, that is, have a higher fitness than do the in-

version homozygotes. Now this heterosis is produced by complexes of

genes carried in the respective chromosomes. By using letters as sym-

bols for genes, the matter may be represented thus: one chromosome

has the genes AiB^iDiEi and the other A2D2C2B2E2. The heterosis

is produced by the interaction of all these genes in the heterozygotes

A1BiCiDiE1/A2D2C2B2E2, whereas the homozygotes which have two

chromosomes with AiBiC]DiEi, or two chromosomes A2D2C2B2E2,

have inferior fitness. The importance of the inversion of a segment

of the chromosome is that the inversion prevents or greatly reduces

the crossing over between the chromosomes (see page 144), and conse-

quently the recombination of genes and the breakup of the gene com-

plexes which produce the hybrid vigor.

Species as Systems of Genes. Heterosis found in inversion hetero-

zygotes in populations of Drosophila sheds some light on a far more

general problem, that of structure of the gene pool in many or all

organisms which reproduce sexually and by cross-fertilization. Let

us recall (Table 7.1) that a high proportion of the chromosomes in

Drosophila populations produce low fitness when present in double

dose (in homozygous condition), yet natural populations of Drosoph-

Hybrid Breakdown 177

ila consist of individuals most of which are healthy and vigorous.

Human populations behave like those of Drosophila flies: most "nor-

mal" human beings probably carry in heterozygous condition genes

which would produce more or less serious disabilities if they were

allowed to become homozygous. The chances of their becoming

homozygous are increased when relatives marry; this is why the inci-

dence of hereditary diseases is high in small populations in which in-

breeding is common. For most natural populations, at least in com-

mon and successful species, inbreeding, however, is rather rare. Most

of the genes in such populations are represented by numerous different

alleles, any one of which is seldom homozygous. The fitness of the

homozygotes, then, is of relatively little importance to the species.

Far more important for the adaptedness of the population is the fitness

of the heterozygous combinations, since most individuals in sexual

populations are heterozygotes.

The situation just described has an interesting and important conse-

quence: natural selection retains in natural populations those gene

alleles which yield vigorous heterozygotes with other alleles present

in the same populations. The gene pool of a sexual population comes,

therefore, to consist of genes that are coadapted, that is, that fit well

together when present in heterozygous individuals. To use a simile,

natural selection in sexual organisms encourages genes that are "good

mixers" rather than "rugged individualists."

The process of coadaptation makes a species something more than

a collection of individuals. A sexual population (Mendelian popula-

tion) is a system of genes which fit together, and yield highly fit

individuals in the environments in which the population lives.

Hybrid Breakdown. Natural selection eliminates from populations

of a species the genes which do not have the property of being "good

mixers" with other genes in the same populations. But genes in dif-

ferent species, or even in remote races of the same species, need not

be "good mixers." Thus at least some of the genes of the horse are not

coadapted with the genes of the ass. The hybrids between these spe-

cies, mules, show a disharmony—they are sterile. Even though mules

may be vigorous and resistant animals as individuals, their adaptive

value in the genetic sense is zero; they do not pass their genes to any

progeny. Species of cottons produce vigorous and sometimes fertile

FI hybrids. The F2 hybrids, or the progenies obtained by backcrosses

to the parental species, however, suffer a breakdown of fitness. Such

hybrids are mostly weak and stunted "rogues" which cannot compete

178 Species

with their parents either as wild or as agricultural plants (see more

about cotton hybrids in Chapter 9).

Not only hybrids between species but even those between remote

races may suffer genetic breakdown. Wallace, Vetukhov, and Brncic

have recently found that when populations of Drosophila pseudoob-

scura which live in the states of California, Nevada, Utah, and Colo-

rado are intercrossed there is a perceptible loss of fitness among the

FZ hybrid progenies. Moore has shown (1946 and later) that when

leopard frogs (Rana pipiens) from the northern part of the United

States (for example, from New England) are crossed to frogs of the

same species from Florida or from Texas the hybrid embryos die be-

fore completing their development. The northern and the southern

leopard frogs behave as though they belonged to different species;

they show in Moore's experiments a complete reproductive isolation

because of the inviability of their hybrids. These northern and south-

ern races are nevertheless capable of exchanging genes through the con-

necting populations which live in geographically intermediate areas.

Thus the leopard frogs from New Jersey or from Oklahoma produce

normally viable, or only slightly abnormal hybrids, when crossed to

Vermont frogs. It is, then, perfectly conceivable that a useful gene,

or gene combination, arising in Texas may eventually reach the Ver-

mont populations, and vice versa.

Different species have different genes because they are adapted to

live in different environments. Each species occupies a certain eco-

logical niche in the economy of nature. Thus Drosophila pseudo-

obscura lives in warmer and drier habitats than Drosophila persimilis

(see above). The genetic system of every species makes it fit to

occupy its ecological niche and to lead a certain mode of life. The

low fitness of interspecific hybrids, therefore, is a by-product of the

parental species' becoming adapted to different environments. Gene

exchanges between populations of distinct species are likely to produce

genotypes which are adapted to neither of the environments to which

their parents are adapted. As the popular saying goes, they are likely

"to fall between two stools."

The harmful results of gene exchange between species, however,

should not be exaggerated. Some of the combinations of genes which

arise owing to hybridization of species may actually be valuable, since

they may be adapted to environments to which neither parent is

adapted. In the state of nature the adaptive value of mules would

be zero, on account of their sterility; but this limitation does not pre-

vent mules from being very useful as farm animals. Some of the

Origin of Reproductive Isolation between Species 179

most important cultivated plants are descendants of species hybrids

(Chapter 9). Cultivated plants and animals live in man-created en-

vironments, which are often greatly different from natural environ-

ments. To fit his crops and his herds to the new environments, man

had to provide them with new genotypes adapted to these environ-

ments. Such genotypes were produced in most instances by unre-

lenting artificial selection within a single species, but in some instances

also by selection among hybrids between species. In doing so, man

merely followed the methods of nature, which occasionally makes use

of introgressive hybridization (page 127) to contrive adaptations to

certain environments.

Origin of Reproductive Isolation between Species. Hybridization

of the northern and the southern races of the leopard frog resulted,

in Moore's experiments, in the production of inviable hybrids (see

above). Such hybrids, however, are not produced in nature since

the frog races involved are allopatric; the Vermont frogs have no

chance to meet Florida frogs except in the laboratory of a biologist.

The diffusion of genes between these races takes place only through

the geographically intermediate races, which yield viable hybrids

with their neighbors.

Compare with this the situation of the two Drosophila species—

Drosophila pseudoobscura and Drosophila persimilis (see pages 170-

171). They live side by side in a territory extending from British

Columbia to California (Figure 8.1), and in this whole territory they

can meet, mate, and beget hybrid offspring. The hybrid males are

sterile, and the backcross progenies suffer a hybrid breakdown, and

are weaklings if they survive at all. The hybrids may be said to be

adaptively worthless, at least in the state of nature. Yet when hybrids

are produced, they consume food and occupy the place in the sun

which can otherwise be exploited by the parental species. Building

up reproductive isolating mechanisms that would prevent the hy-

bridization of the species and the production of the hybrids should,

then, be promoted by natural selection.

An experimental demonstration that selection can, indeed, build up

reproductive isolation of species has been given by Koopman (1950).

We have seen (Table 8.1) that one of the mechanisms which keep

the species Drosophila pseudoobscura and Drosophila persimilis apart

is sexual isolation. The flies show a preference for mating within the

species and an aversion for mating between species. For reasons

that are obscure, this sexual isolation is more nearly complete in the

natural habitats of the flies than it is under laboratory conditions. In

180 Species

particular, the sexual isolation is weakened in the laboratory at low

temperatures. If a mixture of equal proportions of the two species

is kept at 15°C (59°F), the copulations between species are only

slightly less frequent than those within the species.

Koopman made mixed populations of the two species, and main-

tained them for several generations. The experiment was so arranged

that the hybrid flies that were produced were, in every generation,

picked out and discarded. The non-hybrid flies, which were pro-

duced evidently by intraspecific matings, served to propagate the

populations further. The result of the experiment was that the pro-

portions of the hybrids produced in the populations waned from gen-

eration to generation; the proportions of the.matings within the species

gradually increased, and the proportions of interspecific copulations

dwindled. The interpretation of these experimental results is simple.

The original populations of one or of both species were evidently not

uniform with respect to their sexual preferences. Some flies in ex-

perimental environments were more prone to mate with individuals

of the foreign species than were other flies. Since the hybrids were

eliminated from the populations, the flies which mated only with those

of their own species contributed more offspring to the following gen-

erations than the more promiscuous individuals. This situation caused

a process of selection of the more discriminating variants, and the out-

come was an intensification of the sexual isolation. It is a reasonable

working hypothesis that some of the isolating mechanisms between

species are perfected also in nature by a process of natural selection.

Static and Dynamic Approaches to the Study of Species. The con-

cept of species was developed in biology with a practical end in view.

Linnaeus and his successors described and named species of animals

and plants and built a classification of living beings.' This work of

systematization and classification is obviously the foundation of scien-

tific biology. We must know, and be able to tell others, with just

what creatures we are working. Linnaeus's work is being continued

now by numerous investigators on the staffs of zoological and botanical

museums in all civilized countries, as well as by individual specialists.

When an insect pest is found to injure the crops, or an organism is

suspected of being involved in the distribution of a disease, or simply

when anyone finds an interesting creature, it is first of all sent to a

specialist to be "determined," that is, to have its species name attached

to it. Living and fossil species and races are so numerous that nobody

knows all or even a large part of them; the work of systematics has

necessarily to be done by specialists.

Rigid Definition of Species Is Impossible 181

It is not surprising that for the practical work of naming and label-

ing species of animals and plants it is convenient to treat these species,

so far as possible, as completely discrete pigeonholes in which to place

museum specimens. When subspecies or races are described, it is

expedient to treat them also as discrete pigeonholes. This is why

Linnaeus espoused the view that species are created entities (page

166). But the knowledge of species as biological phenomena has not

stood still since Linnaeus. Lamarck and Darwin, and since Darwin

a great majority of biologists, regard species as branches of the stream

of evolution. These branches become subdivided further and further

when populations of a species living in different countries differen-

tiate genetically and become different races. Diverging geographic

races become incipient species when they begin to develop reproduc-

tive isolation. The speciation becomes consummated and irreversible

when the reproductive isolation is complete.

Of course modern systematists not only accept the evolutionary

view of species but also contribute greatly to the study and the under-

standing of the process of speciation. There is, nevertheless, some

dualism in the usage of the species concept in biology. To a practical

systematist whose task is to label species and races, intermediate

specimens transitional between these species and races are obviously

a nuisance. It is hard to assign names to intermediates. By contrast,

a biologist who studies the process of speciation will look for just such

transitions, for races which are about to become species or for in-

cipient species which have not yet completed the development of the

reproductive barriers between them. As a result, an investigator who

gives names to species and races is likely to prefer a more static species

concept, whereas one who studies the process of evolution will incline

towards a more dynamic view. This explains the fact which seems

otherwise very odd: nobody has yet proposed a definition of what

constitutes a species that would be satisfactory to all biologists.

Rigid Definition of Species Is Impossible. In the history of biology

attempts have been made again and again to give a definition of

species with the aid of which we could always decide whether certain

groups of organisms are races of the same species or are distinct

species. For example, it has repeatedly been suggested that forms

which produce inviable or sterile hybrids when crossed belong to dif-

ferent species, whereas those which produce fertile hybrids belong to

the same species. Such a definition of species is unsatisfactory on

two counts. First, the sterility and inviability of hybrids are matters

of degree. Some fairly well-authenticated instances of fertility of

182 Species

female mules are on record, but it would be entirely unreasonable to

say that horse and ass belong to the same species because one female

mule per thousand is fertile. On the other hand, the Vermont and

Florida races of the leopard frog are races of a single species, despite

the fact that the hybrids between them are inviable. These races are

able to exchange genes by way of geographically intervening popula-

tions. Second, as pointed out above, some very distinct species pro-

duce fertile hybrids in laboratory experiments but do not cross at all

or only rarely in nature. Species may be kept apart by various repro-

ductive isolating mechanisms, not only by hybrid inviability and

sterility.

An even more fundamental reason why species cannot be rigidly

defined is that the process of evolution and of species formation is in

general a slow and gradual one. When we observe the animals and

plants which live in the world around us, we see a single cross-section

in time of the evolutionary family tree, as shown schematically in

Figure 8.2. The branches of this family tree are like cables consisting

of many strands—Mendelian populations which inhabit various locali-

ties. At the time level A strands form a single cable—a species not

differentiated into clear-cut races. At another time level B, the strands

are already segregated into two or more bundles and a few inter-

mediate strands. This is a species broken up into races or incipient

new species. At the time level C, we find separate cables of fully

formed species. At the time levels at which we happen to live, some

species are in the state A, others in the state B, and still others are

groups of species, as in C.

When we observe the situation A, or one intermediate between A

and B, there is no doubt at all that we are dealing with a single spe-

cies. Thus mankind is a single biological species, Homo sapiens, sub-

divided into genetically not very sharply demarcated races which have

not begun to evolve reproductive isolation. The situation C is equally

clear. Nobody doubts that man, chimpanzee, gorilla, and orangutan

are distinct species. So are the domestic cat, lion, tiger, jaguar, and

puma. The situation B, however, is the difficult one. Here the popu-

lations are caught in transition, at the borderline between races and

species. Here Darwin's famous advice is applicable: "In determining

whether a form should be ranked as a species or a variety, the opinion

of naturalists having sound judgment and wide experience seems the

only guide to follow." This is, for example, the case with the "rings of

races" discussed below (page 185).

It may be recalled at this point that the impossibility of drawing an

184 Species

population. Until civilization had developed, human races were also

allopatric (see Chapter 13). At present some human races exist also

sympatrically, as do also races (breeds) of domestic animals and

plants. This became possible because of the rise of social isolating

forces (in man) and of the regulation of the reproduction by human

conscious or unconscious effort (in domesticated forms).

Yet as races diverge more and more, they become adapted to dif-

ferent environments in their respective territories or to different modes

of life. Suppose, for example, that two races of an insect species be-

came adapted to feed on different plants, on pine and on fir trees. In

many places pine trees and fir trees grow together, in mixed stands.

So long as the races are allopatric they cannot fully exploit the avail-

able resources of food. They cannot exist together for long in the

same territory, because when they meet, they mate and produce hybrid

offspring; soon the hybrids would far outnumber the parental races.

What is the escape from this biological dilemma? It is, clearly, the

development of reproductive isolation between the races. A repro-

ductively isolated species specialized to feed on pines, and one spe-

cialized to live on fir trees, can live together wherever both food plants

are available, and separately where only pines or only firs are present.

Species may be either allopatric or sympatric.

The essential feature of the process of speciation, of transformation

of races into species, is, then, the development of reproductive isola-

tion between Mendelian populations. Species are the outcome of this

process; species of sexual and cross-fertilizing organisms may accord-

ingly be defined as reproductively isolated populations or groups of

populations. Man is a single species, because the races of which it is

composed show no reproductive isolation, and are in fact exchanging

genes at increasing rates in many parts of the world, the social and

cultural barriers notwithstanding. Horse and ass are different species,

because they are reproductively almost completely isolated, and the

gene exchange between them, if any, is of negligible importance for

the species as wholes.

Borderlines between Species and Race. The above definitions of

speciation and of species do not permit drawing an absolute distinc-

tion between race and species, even in sexually reproducing organisms.

When a situation like that represented as the level B in Figure 8.2 is

encountered, it would be arbitrary to say whether certain populations

are to be considered races or species. Such situations do occur in

nature. We have considered above the races of the leopard frog

studied by Moore. The races in Vermont and in Florida are repro-

Biological Species and Species of Systematics 187

live together, apparently without mixing. But elsewhere in southern

Mexico the populations are variable mixtures of the two "pure" forms.

If the locality, or localities, where the two towhees live together did

not exist or were not discovered, we would be justified in placing all

these birds in a single species with several races. The existence of

the sympatric and evidently reproductively isolated populations makes

this interpretation inadequate. We may think that this is a case of

relatively recent hybridization of species which until then were be-

having like good species ecologically isolated in different habitats.

The burning of the forests, the cultivation of land, and other man-made

changes in the environment have created new habitats, in which both

towhees can live and in which they are no longer isolated.

Biological Species and Species of Systematics. We have chosen to

regard the development of reproductive isolation between diverging

races as the touchstone of species formation. This is not an arbitrary

choice. As pointed out above, the attainment of reproductive isolation

is a very significant stage in the evolutionary path of a population. It

transforms genetically open systems (races) into genetically closed

systems (species); it makes the evolutionary divergence of populations

irreversible; it puts, so to speak, a seal of finality on the populations

being committed to different ways of life.

To discover directly whether certain populations are or are not

reproductively isolated, experiments must be made to test whether or

not various isolating mechanisms discussed above are present or ab-

sent. Zoological and botanical systematists who describe species and

races of am'mals and plants are, however, very seldom in a position

to make such experiments. Indeed, systematists work mostly wiHi

dead and preserved specimens stored in museums and herbaria. Most

species are described by people who never had a chance either to

see them alive or to visit the territories where these species occur.

How can systematists find out whether the organisms which they study

are or are not reproductively isolated? Fortunately, this can be done

in many instances. The methods of study used by modern systematists

yield indirect but usually reliable evidence on this score.

Systematists regard as species groups of creatures which are clearly

different in body structure, and among which transitional or interme-

diate specimens are absent. Domestic cat, lion, tiger, jaguar, and

puma have always been considered species and not races, because no

specimen has ever been found which could not be assigned to one of

these species with certainty. The absence of intermediates between

populations is a presumptive evidence that these populations are

188 Species

reproductively isolated. If they were not isolated and exchanged

genes, intermediates probably would appear. Thus intermediates be-

tween even the most distinctive human races are known. If the popu-

lations live in the same country, sympatrically, and yet intermediates

between them do not occur, they are almost certainly isolated repro-

ductively. We must, of course, beware of mistaking the clear-cut

variants within a polymorphic population for distinct species (see

Chapter 7). Intermediates between the different blood groups in

man are absent, but this does not make the carriers of these blood

groups belong to different species. Mistakes of this sort have been

made from time to time, and this is why Darwin considered "sound

judgment and wide experience" necessary in those who study species.

A more serious source of difficulty in recognizing species is the

existence of distinctive allopatric forms. Sympatric species produce

no intermediates because they are reproductively isolated. But when

populations live in different countries, allopatrically, and testing them

experimentally is not practicable, the decision is often hard to reach.

Most often the systematist resorts to studying the populations of geo-

graphically intervening territories; if these populations are interme-

diate between those found in remote places, there is probably some

gene exchange between all the populations, and they belong to a

single species. This is why we could, for example, infer that man is

a single species even if we did not know directly that there is no

reproductive isolation between any human populations. But what to

do with the sycamore, Platanus occidentalis and the plane tree, Plat-

amis orientalis, which are native respectively to eastern United States

and to southeastern Europe and western Asia? They are quite dis-

tinctive in appearance, but they can be crossed artificially and pro-

duce vigorous and fertile hybrids. No intermediates between them

occur in nature, but such intermediates have little chance of being

formed since the geographic distributions of the two kinds of trees are

widely separated, and the intervening countries have no native plane

trees at all. Botanists regard them as distinct species because they

look so different, and the geneticist Stebbins agrees because they would

probably be kept apart by ecological isolation if they were brought

together in the same territory.

Sibling Species and Species in Asexual Organisms. The reproduc-

tive isolation between Drosophila pseudoobscura and Drosophila per-

similis has been discussed above (pages 170-171). There can be no

doubt that biologically they are distinct species, yet these flies are so

similar in external appearance that a zoologist who studies them

Sibling Species and Species in Asexual Organisms 189

under a small magnification of a microscope cannot distinguish them

at all (they do differ in a more recondite way—the genitalic armatures

of the males show a slight but constant difference). Species which

are clearly distinct biologically but are very similar or identical in

appearance are referred to as sibling species. Their existence may be

troublesome to systematists working with insects preserved on pins in

museum drawers, but it is very enlightening to the science of biology

as a whole.

When we observe that two organisms are different in appearance, it

should be kept in mind that externally visible differences are only the

outward signs of physiological differences, which are in turn based on

genetic differences. It happens that in most living creatures genetic

differences are reflected in externally visible distinctions. This fact

validates the methods of study of species used by systematists. Their

species and biological species are usually the same things. But in

some organisms, such as the two Drosophila species discussed above,

genetic differences great enough to give rise to reproductive isolation

have not produced visible changes in external body structures. Con-

sidered biologically, these species deserve being recognized as such

no less than the tiger and the lion, which happen to be easily dis-

tinguishable in external appearance. To say that insect species must

necessarily be distinguishable in museum specimens, pinned and dried

out as has been customary with insect collectors for two centuries, is

like asking modern physicians to use only the instruments and drugs

which physicians used two hundred years ago.

The definition of species as reproductively isolated populations has,

however, no meaning where sexuality is lost or where self-fertilization

is the usual or exclusive method of begetting progeny. Yet bacteriolo-

gists have to describe bacteria which reproduce chiefly asexually and

form clones, and botanists have to deal with plants like wheats most

of which reproduce chiefly by self-pollination and form pure lines.

Since Linnaeus, it has been customary to give species names to all

organisms, regardless of the methods by which they reproduce (and

these methods of reproduction have become understood only recently

anyway). The clones of bacteria and the pure lines of wheats do form

assemblages of similar forms adapted to live in similar ways; these

assemblages are called species. This is a reasonable procedure; we

should only keep in mind that the "species" in asexual and in self-

fertilizing organisms are not the same biological phenomena as species

of insects, birds, mammals, or of cross-fertilizing higher plants.

190 Species

Are Species Arbitrary Groupings or Natural Entities? The world

around us is infinitely variable. We never encounter twice the same

situation, and objects which we observe, whether living or inanimate,

change with time. To make this infinite variety intelligible and man-

ageable, the human mind combines the situations and objects which it

regards as similar in some respects into groups, and gives to each

grouping a name. Thus there is a class of objects called furniture;

within this class we may distinguish tables, chairs, shelves, etc.; among

tables, office tables, dining-room tables, kitchen tables, etc., can be

distinguished. Living organisms are a part of our environment; they

are enormously diversified, and they have to be classified in order to

be comprehensible. The species has served, and is serving, as a con-

venient category of classification.

However, sexual reproduction makes species of sexual organisms

something more than a tool for classification. Individuals of sexual

species are interdependent in a way analogous to cells in a multicellu-

lar body. Cells are physiologically, and individuals of sexual species

are reproductively, interrelated. The reproductive interdependence

of members of sexual species is necessary for continued existence of

their kind, generation after generation. Thus mankind is more than a

name for a class of objects. It is a biological reality, not only because

all men are kin to each other by virtue of common descent, but even

more so because all human beings are, at least potentially, part and

parcel of the gene pool from which the genetic endowments of then-

posterity will arise. A sexual species is the ultimate extension of the

ancestral family; it is also the family of the future.

Suggestions for Further Reading

The references given at the end of Chapter 7 will be useful for a study of the

problem of species. The problems of race and species are closely interrelated.

Evolution under Domestication

and Evolution by Polyploidy

When The Origin of Species was published in 1859, neither Darwin

nor any one else claimed to have witnessed one species giving rise to

another. Darwin's theory was a scientifically legitimate inference

from indirect evidence. Numerous facts in biology made sense on

the hypothesis that species evolvp by gradual change and differentia-

tion from races. Nevertheless, Darwin collected with great care the

observed facts which showed that organisms undergo genetic changes

and may become altered in the course of time. He found many such

facts recorded by the breeders of domesticated animals and plants.

The breeders produce new varieties of animals and plants to suit

their needs or fancies. Moreover, they do so by means of artificial

selection, often following hybridization of diverse varieties or races.

Darwin's natural selection was a counterpart of artificial selection

which had been known to be effective in domesticated forms. In 1868,

Darwin summarized the evidence concerning evolution under domesti-

cation in a book entitled The Variation of Animals and Plants under

Domestication.

Experimental evidence of the occurrence of evolution is at present

much greater than it was in Darwin's time. Among microorganisms,

evolutionary changes due to interaction of mutation and selection can

be observed under controlled experimental conditions (see Chapter

5). To be sure, these changes involve alterations of single genes or,

at most, of small numbers of genes. They are called microevolutionary

changes, to distinguish them from macroevolution, which results in

production of new genera, families, and classes. Macroevolution en-

tails changes in many, perhaps in all, genes composing the genotype.

Evolution under domestication continues to be interesting to modern

191

192 Evolution under Domestication

evolutionists; it involves genetic changes of a magnitude intermediate

between microevolution, which is easily observable in a test tube, and

macroevolution which requires time on a geological scale.

A special kind of experimentally observable evolutionary change is

caused by the doubling of the chromosome complement—by polyploicly

(see pages 201-208). The polyploids often behave with respect to

their progenitors as full-flsdged new species. Here we have, then,

experimental origination of new species—a feat which was beyond

reach of biologists in Darwin's time. Moreover, the production of

new species by polyploidy is known to occur not only in experiments

but in nature as well. Experimental demonstration of the origin of

species was completed when some naturally occurring polyploid spe-

cies were resynthesized from other species in experiments.

Degree and Antiquity of Domestication. There is no generally ac-

cepted definition of what a "domesticated" animal or plant is. For

our purposes it will be satisfactory to regard as domesticated those

forms which regularly reproduce in captivity and whose populations

are controlled by man. It is man who determines how much wheat or

corn is to be sown, and which kittens or puppies are to be kept and

which disposed of. Even with this definition it is hard to tell how

many animal and plant species have been domesticated. Some forms

have been so thoroughly domesticated that they are utterly unable to

exist except with man's help. In corn (maize), the seeds are firmly

anchored in the cob, and the whole ear is enclosed in a protective

cover (the husks) which prevents the seeds' reaching the soil. No

wild plant with such an arrangement of seeds could possibly survive,

and, as will be shown below, the problem of the origin of corn is not

an easy one. On the other hand, the rubber tree (Hevea brasiliensis)

grows wild in the jungles of equatorial Brazil, as well as on planta-

tions in tropical Asia. The wild and the planted populations of the

rubber trees are only beginning to diverge genetically. The domesti-

cation of the dog took place 10,000 to 12,000 years ago, in the Middle

Stone Age (Mesolithic); that of the maral deer in Asia, and of the

little Australian parrot budgerigar only a century ago. Domesticated

forms of some species were obtained and then lost. Thus in ancient

Egypt, about 3000 to 1000 B.C., there were domesticated varieties of

the Nilotic goose and of at least two species of antelopes, which now

exist only as wild species. Some authorities believe that the inhabit-

ants of southern Chile had, in pre-Columbian days, domesticated the

giant sloth, a huge animal at present known only as a fossil; its bones

as well as remains of its skin, hair, and dung have been found.

Degree and Antiquity of Domestication

193

TABLE 9.1

PLACE AND APPROXIMATE DATE OF DOMESTICATION OF PRINCIPAL

DOMESTIC ANIMALS

(After Krumbiegel, modified.)

Number of

Common

Domestic

Name

Scientific Name

Place of Domestication

Date of Domestication

Varieties

Dog

Cam's /am tJ tans

Northern Hemisphere

10,000-8000 B.C.

200

Cat

Felis maniculata

Egypt

' 2000 B.C.

25

Cattle

Bos taunts

Europe, western Asia

Neolithic, 6000-2000 B.C.

60

Water buffalo

Bos bubalus

India

Prehistoric

T

Bali cattle

Biboa sondaicus

Indonesia

Prehistoric

?

Yak

Poephagus grunnifns

Tibet, Turkestan

Prehistoric

r

Sheep

Ovis aries

Europe, middle Asia

Mesolithic and Neolithic

so

Goat

Capra hircus

Europe, middle Asia

Mesolithic and Neolithic

20

Maral

Cervus moral

Southern Siberia

A.D. 1850

i

Pig

Sus scrofa

Europe, western Asia

Neolithic, 5000-2000 B.C.

85

Horse

Equus caballus

Europe, Asia

Neolithic and Bronze,

60

3000-2000 B.C.

194 Evolution under Domestication

to believe that they all belong to the same species. Compare the

little Chihuahua with a Great Dane; or an ebullient Fox Terrier with

a stolid Saint Bernard. Nevertheless, all these breeds do belong to

one species, since they are capable of exchanging genes, and often

do so, either directly by occasional hybridization, or via the interme-

diate breeds where crossing is impossible because of an extreme dif-

ference in size. The dog species has a common gene pool, subdivided,

to be sure, into the gene pools of the various breeds. Domestic

chickens and pigeons are the most ancient domesticated birds, and

they also form numerous breeds. The domestic goose is an exception

to the rule: its domestication is undoubtedly ancient, but it has given

rise to relatively few distinct breeds.

Centers of Origin of Domesticated Plants. The number of plant

species in various stages of domestication is conservatively estimated

at three hundred, which is many times greater than the number of

species of domestic animals. Pioneer studies concerning the origins

of cultivated plants were made in the last century by A. de Candolle

(1778-1841) and Darwin, and in the current century by Vavilov

(1887-1942). Vavilov came to the conclusion that most species of

cultivated plants were first domesticated in a relatively small number

of countries. These countries are called the centers of origin; their

location is shown on the map in Figure 9.1. It is not surprising that

the centers of origin of cultivated plants largely coincide with the

ancient centers of human civilization.

Vavilov has distinguished eight or nine such centers. (1) Chinese,

including China, particularly its mountainous interior, and Korea.

This center has given mankind many varieties of oats, barleys, mil-

lets, soya and other beans, bamboo, sugar cane, some cabbages, oranges

and lemons, peaches, apricots, pears, prunes, cherries, tea, poppy, and

many other useful plants less well known to our Western civilization.

(2) and (3) Indian and Indomalayan centers, which may also be

treated as a single one, embracing India (but not western Pakistan),

Burma, Indochina, and western Indonesia. This is the land of the

origin of rice, which is estimated to be the main staple to perhaps

60 per cent of humanity, sorghum (Indian), some beans, eggplant,

cucumber, bamboos, sugar cane (Indomalayan), bananas (Indoma-

layan), mango (Indian), some oranges and lemons, breadfruit (Indo-

malayan ), cocoanut palm (Indomalayan), diploid cotton, pepper (In-

dian) and many less widely known plants. (4) Middle-Asiatic, in

northern Pakistan, Kashmir, Afghanistan, in the southern part of Rus-

196 Evolution under Domestication

sian Middle Asia (Turkestan). Here are found most soft (bread)

wheats, some beans, melons, onions, spinach, some apricots, apples,

pears, almonds, grapes, and walnuts. (5) Western Asiatic, in the in-

terior of Asia Minor, Transcaucasia, and northern Iran. Here are

native some varieties of hard and soft wheats, rye, oats, lentils, peas

flax, poppy, melons, carrots, and some varieties of pears, cherries, al-

monds, grapes, figs, and other fruits. (6) Mediterranean, countries

around the Mediterranean Sea. This center gave some varieties of

wheats (relatively little used at present), peas, flax, sugar beets, cab-

bages, asparagus, and olives. (7) Ethiopian, in Ethiopia and Eritrea.

This is the center of hard wheats, many varieties of barleys, sorghum,

flax, sesame, and coffee. (8) Central American, embracing southern

Mexico, Guatemala, and possibly some of the West Indian Islands.

This is the source of such immensely important cultures as corn

(maize), some beans, sweet potatoes, upland cottons, peppers, papaya,

and many other tropical fruit trees. (9) South American or Andean,

in Peru and adjacent parts of Equador and Bolivia. This region

shares the credit for having given origin to corn and to some of the

most useful varieties of cottons, and also to potatoes, tomatoes, to-

bacco, pumpkins, and the quinine tree. Somewhere to the east

of the Andean center, in tropical Brazil, pineapples, cacao, and the

rubber tree originated.

It is hard to estimate the importance to mankind of the achieve-

ments of the primitive agriculturists in the above lands. They man-

aged the critical first steps of the evolution under domestication of

most of the plant species which now nourish mankind. They did this

immensely difficult work without the benefit of modern scientific

knowledge or of modern technology.

Wild Relatives of the Domestic Horse. We choose for a more de-

taile.d consideration of evolution under domestication one animal

species, the horse, and two plant species, cotton and corn. These three

species will exemplify different kinds of evolutionary changes that

occur, and will illustrate the different kinds of uncertainties with

which the research in this field of evolutionary biology has to cope.

The genus Equus, to which belongs the domestic horse (Equus

caballus), contains three groups of species now living in the wild

state. One of the groups is the zebras, which comprise three species

with numerous races living in eastern and southern Africa. Another

group consists of the asses, of which two species are recognized—the

onager (Eqnus hemionus) in the steppes of middle Asia, and the true

Wild Relatives of the Domestic Horse 197

wild ass (Equus asinus) in northeastern Africa. The latter is without

doubt the wild progenitor of the domestic donkey.

Neither the zebras nor the asses have a part in the ancestry of the

domestic horse. Domestic horses and domestic asses are crossed on a

large scale to produce mules. In 1920 there were about 5,650,000,

and in 1949 about 2,353,000 mules on farms in the United States

alone. The practice of breeding mules is an ancient one, well known

to Aristotle. But mules are interspecific hybrids which are almost

invariably sterile; the few recorded fertile female mules certainly do

not provide a channel for a regular gene exchange between the gene

pools of these species (see Chapter 8). Zebras have been crossed to

horses in zoological gardens, but the hybrids, called zebroids, are

sterile. Zebras, asses, and horses are close relatives which arose rather

recently, geologically speaking, from a common ancestor (see Chap-

ter 12). But they have diverged genetically far enough to become

reproductively completely isolated. Their evolutionary paths are

separate.

The only true wild horse lives in a part of middle Asia, more pre-

cisely in the deserts of western Mongolia. It has been given a spe-

cific name of Przevalsky horse (Equus przevalskii)— which is even

more difficult for English speakers to pronounce than is the name of

the author of this book. The wild horse is only as tall as a small pony,

12 to 15 "hands" (48 to 52 inches) from the ground to the withers,

and rather ungainly in appearance from the standpoint of a lover of

horseflesh. It should be noted that it has a short erect mane, no fore-

lock, and is a dun color (cf. page 38). It is wild and intractable even

after a prolonged life in captivity. The wild and the domestic horses

cross easily and produce fertile progeny.

The wild Przevalsky horse is the last survivor of a species which in

historical times was much more widespread. A wild horse called

tarpan was quite common in the steppes of southern and southeastern

Russia as recently as the last century. Tarpans were a nuisance to

the pioneer settlers, since they interfered with the crops and led away

domestic mares. They were hunted and exterminated, the last tarpan

having been killed around I860 in the steppe north of the Black Sea.

The ancient Romans encountered numerous "forest horses" when their

empire spread to what are at present Spain, France, and Germany.

How different the tarpans and the forest horses were from the Przeval-

sky horse is not exactly known. The tarpan was allegedly mouse

colored instead of dun; this difference may be due to a s:ngle gene

(see page 38). The forest horse was relatively tall (about 60 inches),

198 Evolution under Domestication

with stout limbs and a long but narrow head. Its portrait, drawn by

an artist of the Stone Age on a cave wall at Altamira, Spain, shows

an erect mane, like that of the now living wild horse.

Is the Domestic Horse Monophyletic or Polyphyletic? Just when

and where the first horses were domesticated is unknown. The first

historical records of domestic horses appear at almost the opposite

ends of Asia, in Mesopotamia (Iraq) and in China, about simul-

taneously, around 2000 B.C. Horses were brought to Mesopotamia

from the north, possibly from the Russian steppe. It is, hence, con-

jectured that the tarpan was domesticated by the nomad inhabitants

of the steppe a few centuries earlier, possibly around 2500 B.C. But

the horses that came to China were most likely the descendants of the

Przevalsky horse, domesticated in the wilds of Mongolia quite inde-

pendently from the tarpan.

It is most likely, though obviously improvable, that horses were

domesticated many times and in many places in the steppes and

deserts of eastern Europe and Asia, and in the forest zone of western

Europe. The first record of a horse hitched to a chariot appears in

Greece around 1700 B.C., in Egypt 1600 B.C., in India 1500 B.C. By

1000 B.C. horses were known and widely used in most parts of Europe,

and in Asia and Africa outside the tropical forest zones. Mounted

horsemen first participated in the Olympic games in 648 B.C.

The problem of whether the now living domestic horses are de-

scended from a single wild ancestor (monophyletically) or from two

or several ancestors (polyphyletically) has been long and incon-

clusively discussed by the students of the horse origins. Perhaps it

does not really matter. If the gene pool of the domestic horse con-

tained genes of two different species, such as horse and ass, every

one would agree with the polyphyletic hypothesis. But the Przeval-

sky horse, the tarpan, and the forest horse were clearly geographic

races of the same wild species, which at one time was distributed

perhaps from the Atlantic coast of Europe to the Pacific coast of Asia.

Most likely this species could be divided not into three but into more

numerous races, and quite possibly most or all of these races con-

tributed their genes to the gene pool of the modern horse. The re-

combination of the genes contributed by these races, plus the genes

which arose by mutation since the domestication, gave rise to the

variety of horse breeds.

The Domestic Horse Breeds. Regardless of how many wild an-

cestral races were domesticated, their descendants underwent consid-

erable genetic changes. These changes took place because the horses

The Domestic Horse Breeds 199

were used for different purposes, and each purpose was best served

by different genotypes in the horse populations. Natural and artificial

selection perpetuated the suitable genotypes and eliminated the un-

suitable ones. The existing diversity of horse breeds is the result.

Nomadic cultures most completely dependent on horses developed

particularly in Mongolia and in Middle Asia. Here horses were, and

still are, used as meat and milk animals, as well as for riding and

for transportation of the tents in which the nomads live and their

furnishings. Horse meat and fermented mare's milk still are im-

portant staples in some parts of Asia. The horse breeds suitable for

these multiple uses must be before all else sturdy, tireless, able to

subsist on scant food, and to resist inclement weather. This type

required relatively little change from the wild horse condition, except

development of some tameness. The horses of Mongolia are the most

"primitive" existing breeds, clearly resembling the wild Przevalsky

horse, including the possession of an erect mane and a high frequency

of the dun coat color. The hordes of Genghis Khan, mounted on

such horses, erupted from Asia into Europe in the thirteenth century.

They swept all resistance before them; this "Scourge of God" could

not have occurred except for a perfect balance between the cultural

heredity of a human population and the biological heredity of a horse

population.

Quite different types of horses appeared in western Europe, where

horses were used and valued primarily for warfare. But here the

mounted warrior tried to make himself less vulnerable by donning

progressively heavier armor. The speed of the horse had to be sac-

rificed to its ability to carry great weights. Hence the Middle Ages

saw the appearance in Europe of large, heavy, and powerful breeds of

horses, perhaps having a strong dose of the genes of the forest horse

which lived wild in these parts.

The invention of firearms, however, ended the usefulness of armor.

The horses which used to carry knights in battle now began to serve

a different purpose: transportation of heavy loads and heavy farm

labor. Thus arose the powerful "cold-blooded" draft horses: Per-

cherons in France, Belgians in Belgium, Clydesdales, Shires, and Suf-

folks in England. Stallions of these breeds are often enormous ani-

mals, 64 to 70 inches tall, weighing 1800, 2000, and even 2200 pounds.

The American farm horses carry some genes derived from these breeds,

together with those of miscellaneous "utility" horses of western Euro-

pean origin.

Still other horse breeds were developed in Mediterranean countries

200 Evolution under Domestication

and in the Near East, also in response to war needs. Until the inven-

tion of saddles with stirrups, the cavalry either fought standing in a

horse-drawn chariot or the men dismounted to begin combat. The

Greeks and Romans did not have stirrups. The stirrups, invented

somewhere in Asia, furnish fulcra which permit mounted combat.

The speed, agility, and alertness of the cavalry horse now became

vital. Accordingly, there appeared various breeds of "light" riding

horses, among which the Barb in northern Africa and the Arabian

were most important. Just when these breeds first appeared is un-

certain, but by the time of Mohammed, in the seventh century A.D.,

Arabia possessed a highly valued variety of horses. The Arabian horse

is regarded by many people as the most beautiful of all horse breeds,

if, indeed, not the most beautiful of all animals. However that may

be, it is certain that the Arabian and related Eastern breeds have been

highly esteemed as riding animals for centuries. Their genes have

become diffused the world over in many more recent breeds.

The Race Horse. If the origins of the breeds discussed above are

lost in the mists of antiquity, that of the English thoroughbred breed

is known in fair detail. After the passing of the heavily armored

medieval knights mounted on slow horses, light but fast Eastern horses

were imported into western Europe at increasingly frequent intervals.

A population of progressively lighter riding horses gradually resulted

from hybridization of the imported and local breeds. A new breed

was started in England by mating several local mares with three

stallions imported from the East between 1689 and 1730. These stal-

lions were either pure Arabian, or at any rate carried many genes

derived from the Arabian breed. The hybrid offspring were mated

among themselves, and in the second generation three stallions were

selected—Matchem, Eclipse, and Herod. Most thoroughbred horses

are descended from these stallions and from their female siblings and

relatives.

The pedigree of the thoroughbreds, carefully recorded in the so-

called stud books, is a most complex network of descent relationships

beginning with the horses named above. All individuals are multiply

related to each other. It should be realized, however, that this pedi-

gree involves also a most intense and careful selection of individuals

possessing the desired properties, and of these the most desired has

been speed. A modern thoroughbred horse is a strikingly different

animal from its Arabian (or near-Arabian) ancestors. It is consid-

erably taller (64 inches or higher), and capable of much greater

speeds—the present record is a mile run in 1 minute and 33% seconds.

Cultivated Cottons 201

A thoroughbred is also an animal requiring the most elaborate care

and feeding, and it is too nervous and high-strung to be used for any

purposes other than running on a race track or crossing to other horse

breeds to "improve" the breed.

To an evolutionist, the thoroughbred horse is an interesting example

of a great change in the direction chosen by man which can be wrought

in an organism by selection within a rather small number of genera-

tions. And if we compare the thoroughbred with the wild horse from

Mongolia we find the two animals certainly as distinct as or more

distinct than are "natural" species of the genus Equus. Why, then,

have no reproductive isolating mechanisms developed to make them

full-fledged species? The answer is necessarily speculative, but it is

reasonable to think that it is because breeds of domesticated animals

are isolated from each other and from their progenitors owing to the

control of their reproduction by man.

Cultivated Cottons. Despite the recent invention of synthetic fibers

(rayon, nylon, etc.), the cotton plant is, and will probably remain, the

chief source of the textiles which clothe humanity. The most ancient

cotton textiles known are dated about 3000 B.C. They have been

found in the excavations of Mohenjo-Daro, among remains of the Indus

civilization, which is one of the three most ancient civilizations in the

world (the other two being Egyptian and Babylonian). Moreover,

the Mohenjo-Daro textiles are obviously made by a skilled craftsman,

indicating that cotton must have first been used at some time prior

to 3000 B.C. Probably quite independently of the Indus Civilization,

cotton was domesticated also in the Americas, since cotton textiles are

found in the pre-Inca graves in Peru. Cotton was used extensively

by many tribes in pre-Columbian days, including some of the tribes

in the southwestern United States.

There is, however, an important difference between the Old World

and the New World cultivated cottons. All cottons native to the Old

World have 26 chromosomes (13 pairs) in their cells (Figure 9.2). All

New World cultivated cottons have twice as many chromosomes (52

or 26 pairs); hence the New World cottons are polyploid or, to be

more exact, tetraploid (see page 65).

How did this difference arise? One of the possibilities is that the

New World cottons arose from a single ancestral species, having 26

chromosomes, simply by reduplication of the chromosome complement

(by autopolyploidy). Another possibility is that two different species,

each having 26 chromosomes, have crossed, and that the duplication

of the chromosome complement took place in the hybrid. If so, the

Hybrids between Species of Cotton 203

cultivated in tropical America) are, as we know, tetraploid, with 52

chromosomes. So is a wild and lintless species which lives on the

Hawaiian Islands (C. tomentosum).

Hybrids between Species of Cotton. By artificial cross-pollination

it is fairly easy to obtain hybrids between species of Gossypium. The

progeny of an American wild diploid species (26 chromosomes),

crossed to an American cultivated tetraploid species (52 chromo-

somes), will have 39 chromosomes, and will be triploid. Such triploid

hybrids are largely sterile, but the behavior of the chromosomes at

meiosis can be observed under the microscope. The cells at meiosis

show about 13 bivalents (that is, 26 chromosomes that have come to-

gether in pairs) and 13 univalents. Next, let us cross an American

cultivated tetraploid to an Old World diploid species. The progeny

is again triploid, 39 chromosomes, largely sterile, and showing ap-

proximately 13 bivalents and 13 univalents at meiosis.

What do the above observations mean? The most reasonable in-

terpretation is that the chromosomal complement of an American

tetraploid cotton (26) contains a set of 13 chromosomes sufficiently

like those of the American diploid cotton species, so that these chromo-

somes pair up and make 13 bivalents at meiosis in the hybrid. By the

same token, the American tetraploid cotton must contain 13 chromo-

somes like those of the Old World species (see Figure 9.2).

The validity of this interpretation can be tested by more experiments.

Let us obtain hybrids between diploid species. Hybrids between dif-

ferent American diploid species are, of course, diploid; they are fertile,

and show 13 bivalents at meiosis. The same is true for hybrids be-

tween different Old World diploid species, but things go differently

when an American diploid species is crossed to an Old World diploid.

The hybrid is more or less sterile, and its 26 chromosomes form few

or no bivalents at meiosis.* The chromosomes of different Old World

species are still so nearly similar that they "recognize" each other and

pair in hybrids. The same is true for chromosomes of different diploid

American species. But the chromosomes of American diploids be-

come in the process of evolution so different from the chromosomes of

the Old World species that they pair no longer.

* The story as here presented is somewhat oversimplified. Actually the chromo-

somal pairing in hybrid meiosis is variable, and in some cells some chromosomes

fail to pair which do pair in other cells of the same hybrid. This is due to the

occurrence of structural changes in the chromosomes in the evolutionary process

(see pages 65, 148). The chromosomal similarity or dissimilarity is not an all-

or-none affair.

204 Evolution under Domestication

When Did the Tetraploid Cottons Appear? The tetraploid Ameri-

can cultivated cottons are allotetraploid, that is, they contain a set of

chromosomes like those still existing in American diploid species, and

one set like those in Old World diploid species. An ancestor of the

cotton plants now growing on a field in Alabama must have been an

FI hybrid between two different species of cottons, one of which must

have resembled cottons now growing wild in America and the other

growing in the Old World. All students of the problem of origin of

cottons now agree that such hybridization must have occurred. They

do not agree nearly so well about the place and the time of the hy-

bridization.

Indeed, in order to produce hybrids, the cotton species which now

live on different continents on the opposite sides of the globe had to

meet somewhere. Wild cottons now grow only in tropical and sub-

tropical lands; intense and prolonged selection was necessary to obtain

cotton varieties capable of being cultivated in warm-temperate cli-

mates, like that of the southern United States. The only place where

the American continent meets the Old World continental mass is the

vicinity of the Bering Strait, in the Arctic. The climate of the Bering

Strait region is far too cold for any cotton to survive. The alterna-

tive is to suppose that seeds of the Old World cottons were some-

how transported to tropical America either from Asia across the whole

wide Pacific Ocean, or from Africa across the Atlantic. Cotton seeds,

however, are rapidly killed by sea water, so that they could not be

transported by ocean currents. They are too heavy to be transported

by wind.

Harland (1939) sought to escape the above dilemma by supposing

that the crossing of the diploid species which gave rise to the Ameri-

can tetraploid cottons took place in Cretaceous or in Eocene times,

many millions of years ago (Table 12.1). Geologists believe that in

those remote times the climates of the whole earth, including the pres-

ent arctic regions, were much warmer than at present. Some also be-

lieved that there may have existed a continent in the place of the

present Pacific Ocean, but most geologists now definitely oppose this

idea. Be that as it may, Harland surmised that the tetraploid cottons

arose very long ago.

On the contrary, Hutchinson, Silow, and Stephens (1947) assume

that the American tetraploids are of relatively very recent origin. The

inhabitants of the South Sea Islands have made many journeys across

the uncharted ocean. Suppose, then, that some of these seafarers

came to the West Coast of South America and brought with them

Raphanobrassica, an Experimental New Species 205

seeds of an Asiatic diploid cotton with spinnable lint. The plants

which grew from these seeds then crossed to a native American diploid

species which grew near by, and gave an allotetraploid with superior

lint quality, which was picked up and cultivated by the Indians.

Neither hypothesis can be considered proven, although recent cyto-

logical and distributional data make Harland's hypothesis, or some

modification of it, more probable. Anyway, the evolution of cotton

under domestication went remarkably fast. As we know, the wild

species of cotton have no spinnable lint, but only useless fuzz on

their seeds. Before cotton textiles were made at Mohenjo-Daro in

3000 B.C. (see above), a considerable amount of selection work must

have been done. Egypt, the economic welfare of which depends at

present largely on its cotton crop, received cotton much later, about

500 B.C. Ancient Egyptian and Babylonian textiles were made of flax

or of wool. The development of the cotton crop in China and in

Middle Asia is even more recent—between A.D. 700 and 1300. This is

probably because cotton is originally a tropical plant which is peren-

nial in habit, that is, grows for several years. Before it could be grown

successfully in temperate lands, annual races (which complete their

development, from seed to seed, within a year) had to be developed.

Such races were developed both in the Old and in the New World. If

the Hutchinson-Silow-Stephens hypothesis is correct, the American

cultivated cottons arose, possibly somewhere in Peru on the Pacific

Coast of South America, perhaps some two thousand years ago. Yet

by the time Columbus landed, there were two distinct cultivated spe-

cies, Gossypium hirsutum and G. barbadense, with numerous varieties.

Some of these varieties were of such high quality that they were intro-

duced in the Old World and largely displaced there the native diploid

cottons. The commercial cotton crop the world over consists at

present predominantly of tetraploids, which are native in tropical

America.

Raphanobrassica, an Experimental New Species. The hypothesis

of allopolyploid origin of cultivated American cottons is made neces-

sary by much experimental and observational evidence. The event

postulated by this hypothesis occurred, however, in the dim past, but

a quite analogous process has actually been observed in experiments

on other plants. The classical example is the new allotetraploid spe-

cies obtained by Karpechenko (1926) by crossing the common cab-

bage (Brassica oleracea) and radish (Raphanus sativus). These plants,

as their names show, belong to different botanical genera, but they

have similar chromosome numbers—18 (9 pairs). Karpechenko ob-

206 Evolution under Domestication

tained diploid cabbage X radish hybrids, which had, as expected, 18

chromosomes, 9 from cabbage and 9 from radish. The cabbage and

radish chromosomes fail to pair at meiosis in this hybrid, and the

hybrid is sterile.

This is a situation observed also in many other, in fact in most,

sterile interspecific hybrids. It was first seen by the Finnish geneticist,

Federley, in 1913 in hybrids between certain species of moths. Owing

to the occurrence in evolution of inversions, translocations, and other

chromosomal changes, the gene arrangement in the chromosomes of

different species becomes so thoroughly "scrambled" that these chromo-

somes no longer "recognize" each other as homologues. The differ-

ence in the chromosome structure becomes manifest particularly at

meiosis in the hybrids, when the chromosomes of the parents should

normally come together in pairs (Chapter 3). Their failure to do so

initiates grave disturbances in the processes of sex cell formation. The

result is that the hybrid, although it may be quite normal and vigor-

ous, fails to produce normal sex cells and is sterile.

The cabbage X radish hybrid, however, was not completely sterile.

Karpechenko obtained from it several seeds and grew a small F2 gen-

eration. These F2 plants proved to be tetraploid; they had 36 chromo-

somes in their cells, 18 chromosomes of cabbage and 18 of radish.

Such seeds arose from ovules and pollen grains in which the meiosis

failed altogether, and came to possess the entire chromosome comple-

ment of the hybrid—9 cabbage and 9 radish chromosomes.

The tetraploid hybrids grew to be tall and vigorous plants, called

Raphanobrassica. Agriculturally this new plant is, alas, useless, since

it happens to have a root like cabbage and foliage like radish. But

it is a very important plant for a biologist, being the first new species

(or genus) produced in experiment. Indeed, the allotetraploid

Raphanobrassica is quite fertile, its meiosis being quite normal with

18 bivalents normally formed. Since it has 18 cabbage and 18 radish

chromosomes in its cells, there is no difficulty in pairing, and 9 cabbage

and 9 radish bivalents arise. Despite its hybrid origin, Raphanobras-

sica breeds true. Moreover, it is reproductively isolated from its own

ancestors—the cabbage and the radish. The Raphanobrassica X cab-

bage and Raphanobrassica X radish hybrids are formed with difficulty;

they are triploid (27 chromosomes), and largely sterile. In short,

Raphanobrassica is a new plant species by any reasonable definition

of species.

The experimental production of new species is an important achieve-

ment of evolutionary biology. Some still surviving anti-evolutionists

Natural AUopolyploids 207

cling to the forlorn hope that all changes observed in experimental

organisms are intraspecific changes. Indeed, a mutant of Drosophila

melanogaster is still a fly and belongs to the same species as its an-

cestors. A thoroughbred horse is still a horse. But Raphanobrassica

is evidently neither a cabbage nor a radish; it is a new and hitherto

unknown organism—Raphanobrassica.

Natural AUopolyploids. During the thirty years since Karpe-

chenko's pioneer experiments, several new species have been synthe-

sized by allopolyploidy. These experiments merely copy a method of

species formation which nature has used on a fairly large scale, at least

in the plant kingdom. As we have seen above, there is strong evi-

dence that the cultivated American species of cotton arose by this

method. Other species of cultivated and wild plants, among them

most wheats, oats, sugar cane, tobacco, potato, coffee, most cultivated

roses, raspberries, and many other useful plants, are allopolyploids.

Allopolyploidy was probably the method of species formation in plant

genera in which the chromosome numbers of the known species form

a series of multiples of some low "basic" number. Thus wheats and

related grasses have species with 14, 28, and 42 chromosomes (the

"basic" number is, then, 7).

Particularly interesting are the experiments in which experimentally

synthesized allopolyploid species are similar to or identical with nat-

urally occurring ones. Here the evolutionary process which has at

some time in the past produced certain "natural" species is repeated

in experiments. Beasley (1942) came close to resynthesizing an

American tetraploid cotton from hybrids of a wild American diploid

(Gossypium thurberi) from Arizona and an African cultivated diploid

(Gossypium arboreum). He obtained a tetraploid hybrid which was

sterile as a male (that is, failed to produce functional pollen grains).

Its meiosis, however, was normal and it set seed freely when pol-

linated by ordinary American tetraploid cotton. The meiosis in these

backcross hybrids was also nearly normal. It is likely that the Arizona

wild cotton is not the American diploid species which is the ancestor

of the tetraploids, but it is close to the true ancestor.

A full success was gained for the first time by the Swedish geneticist

Muntzing (1930). The species Galeopsis tetrahit, of the mint family,

has twice as many chromosomes (32) as the related species, G. speciosa

and G. pubescens (16). Muntzing crossed the latter two species, and

among the offspring of their hybrids selected plants with doubled

chromosome complements. These plants he called "artificial tetrahit,"

since they reproduced the natural species of that name, freely crossed

208 Evolution under Domestication

to the latter, and gave fertile hybrids with normal chromosome pair-

ing. In short, Muntzing has recreated G. tetrahit, a natural species

which arose by a similar process of crossing two other species and

doubling the chromosome complement in the hybrid. Since 1930

several other natural species have been so recreated.

Origin of new species by polyploidy is, of course, not the most

usual method of emergence of species in evolution, although poly-

ploidy has played an important role in certain groups of organisms,

especially in the plant kingdom. The doubling of the chromosome

number is a process which, when it occurs in a species hybrid, may

create a new species almost instantaneously. This is why species

which have arisen by polyploidy may sometimes be reproduced in

experiments. The emergence of new species from races is, as dis-

cussed in Chapter 8, a slow process which takes quasi-geological time

to be completed. Species which arise in that way may have similar

or different chromosome numbers, but they seldom have chromosome

numbers which are multiples of each other. More important than

changes in the numbers are the internal reconstructions which take

place in the chromosomes by mutation, inversion, translocation, dupli-

cation, and deletion of genes.

Corn and Its Relatives. Except in very cold countries and in wet

tropical lowlands, maize or Indian corn (Zea mays) is one of the

principal crops which nourish mankind. The land area annually

planted to corn is estimated at 221 million acres. The origin of this

highly important plant, however, is far from completely known.

As will be shown below, the uncertainties encountered in studies of

corn origins are of a quite different nature from those to be contended

with in the domestic horse and in cotton evolution.

As pointed out above, corn has been so profoundly modified in the

process of adapting it to man's needs that it is utterly unable to propa-

gate itself without human help and to live as a wild plant. Its wild

progenitor must necessarily have been a rather different-looking plant.

Botanists place corn among the Maydeae, a subdivision of the grass

family. Sugar cane and sorghum belong to the same family, but they

are not closely related to corn. A much closer relative is teosinte,

Zea mexicana (or Euchlaena mexicana, since it is considered by some

to belong to a different genus), which grows in parts of southern

Mexico and of Guatemala, mostly as a weed in or near fields of culti-

vated corn. Teosinte is a tall grass with rather broad leaves, though

not as broad as corn leaves. It resembles corn also in having male

flowers (producing pollen) concentrated in a "tassel" at the top of the

210 Evolution under Domestication

vored by artificial selection. This would mean that corn arose some-

where in Central America, where teosinte is now still growing in semi-

wild condition. To be sure, teosinte is not at present used as food;

but its seeds can be "popped" by heat, thus shattering the horny cover

and exposing the more palatable grain.

It now appears, however, that the teosinte hypothesis was a false

clue to the riddle of maize origin. Mangelsdorf and Reeves suggested

in 1939 that teosinte, far from being the ancestor of corn, is derived

from corn by means of hybridization of primitive corn with a species

of Tripsacum. According to this view, teosinte is in fact genetically

a corn with some sections of Tripsacum chromosomes included in its

genotype.

The evidence that this extraordinary situation actually obtains in

reality comes chiefly from experiments of crossing teosinte with mod-

ern cultivated corn. The hybrids are fertile, and the segregation in

FZ and later generations shows that the genes which determine the

visible differences between corn and teosinte are concentrated mainly

in four chromosome segments in as many chromosome pairs. Both

maize and teosinte have 10 pairs of chromosomes, and it is most un-

usual for species of either plants or animals to have most chromosomes

remain similar in the process of evolution, the changes being confined

to only a few segments of chromosomes. These blocks of genes are

supposed to be acquired from Tripsacum through hybridization which

took place at some unknown time in the past, presumably in Central

America where corn, Tripsacum, and teosinte often occur together.

Despite the striking differences in appearance between corn and teo-

sinte, Mangelsdorf and Reeves propose to consider the latter not a

separate genus, but only a species or even a variety of corn.

Primitive Corn. If teosinte is not the wild ancestor of corn, then

what was this ancestor like and where did it live? This problem must

be regarded as still unsolved, but Mangelsdorf and his collaborators

have obtained interesting data which suggest a probable solution.

More than a century ago, Saint-Hilaire (1829) surmised that corn

might have developed from the so-called pod corn (Figure 9.4). The

chief characteristics of the pod corn are produced by alleles of a

single dominant gene, called tunicate, the presence of which causes

rather drastic changes in the body of the plant. The seeds of pod

corn are each enclosed in a papery cover consisting of enlarged glumes

(the same structures which give the horny covers of the teosinte and

gama grass seeds, see above). The ears on the sides of the main stalk

of the plant are suppressed, or are elongated, branched, and protrud-

Cultivated Corn 213

This does not mean, of course, that corn was domesticated in New

Mexico. Even in the very early times Indian tribes had trade rela-

tions with their neighbors, so that a useful plant or a useful technique

could gradually spread throughout the Americas. One of the guesses

is that the wild ancestor of corn was a rather rare plant growing in a

limited territory somewhere in the interior of the continent of South

America, but this is only a guess. Archeologists may discover in

South America remains of maize as old as or older than that found in

New Mexico, and such a find may make the story of corn much clearer

than it is now.

Cultivated Corn. The spike of the primitive corn found in the

New Mexico cave (Figure 9.5) is a dwarf compared to modern corn.

Surely, the yields which prehistoric farmers obtained from their plant-

ings would be regarded as pitifully small by farmers in our day. The

tremendous increase in yields is one measure of the progress accom-

plished through persistent and long-continued selection since corn be-

came a cultivated plant. We should keep in mind that selection had

to produce not just one high-yielding variety but many varieties,

adapted to different climatic and agricultural conditions in the lands

where corn is cultivated. Some of these varieties are illustrated in

Figure 9.6. A variety that is good in Iowa may be worthless in Peru,

and vice versa. The tropical varieties are genetically adapted to the

short days, and the temperate land varieties to the long summer days

which occur when the fields are normally planted. They simply fail

to flower when exposed to days and nights of a duration abnormal

for them. As a matter of fact, corn is so sensitive to the environmental

conditions that for optimal yields we have to breed genotypes "tail-

ored" to each particular region in which they are to be grown. This

is why separate corn-breeding programs are pursued not only in dif-

ferent countries but also in every state of the United States where

corn culture is economically important.

Mangelsdorf, Anderson, and others suppose that some of the genetic

building blocks for the creation of the varieties adapted to so many

different local conditions may have been furnished by hybridization of

the domestic corn with wild species of Tripsacum. Although Tripsa-

cum is a grass very sharply different from corn in almost every char-

acter, they can be crossed. The hybrids are largely sterile, but some

seeds can be obtained by pollinating the hybrid ovules with corn

pollen. There may occur an introgression (page 127) of Tripsacum

genes into corn populations. As we have seen above, teosinte may

Hybrid Corn 215

(1877) that plant progenies obtained by self-pollination within a

flower, by cross-pollination of different flowers of the same plant, or

of different individuals of the same strain, are deficient in vigor. They

suffer inbreeding degeneration. Conversely, the progenies obtained

by crossing different strains exhibit hybrid vigor (or heterosis, as it is

now called).

In 1908 and 1909 the geneticist G. H. Shull published his studies on

inbreeding and crossing in corn. A corn plant can be "selfed" by

transferring the pollen from the tassel to the stigmas (silks) of the

same individual. In a cornfield most seeds come, however, from cross-

pollination of silks of one individual by pollen grains of other indi-

viduals. The "normal" vigor of the field "variety" is maintained by

the cross-pollination. Shull found that inbred lines, obtained by sys-

tematic selfing, rapidly dwindle in vigor, size, and yield of the plants.

Intercrossing different inbred lines gives progenies in which the vigor

is restored up to the average level of the variety from which the inbred

lines were obtained. But hybrids between some inbred lines may

even exceed the original variety in productiveness.

Here, then, was a "theoretical" investigation which resulted in a

first-rate "practical" discovery. Before this discovery could be ex-

ploited in practical farming, however, another problem had to be

solved. The hybrid seed must be obtained from inbred lines which

consist of weak plants giving poor yields. Getting enough seed from

them to plant large fields is prohibitively expensive. Jones (1917)

solved the problem by means of the so-called double-cross method

(Figure 9.7). Instead of two inbred lines, four inbred ones are used.

They are first intercrossed in pairs, A X B and C X D. The two hy-

brids obtained are vigorous and high-yielding. They are interplanted

in parallel rows, and the tassels are removed from one of the hybrids,

so that its silks can receive pollen only from the other hybrid (A X

B) X (C X D). The resulting seeds are planted on ordinary farms

and grow into high-yielding plants. However, the very high yields

occur in only one generation, and the fields have to be replanted every

year with fresh hybrid seeds.

Corn-breeding programs based on hybrid corn were started soon

after 1917, and by 1933 hybrid corn acquired a considerable importance

in commercial plantings. The average yield of corn on farms in the

United States in the early nineteen-thirties was estimated at about 22

bushels per acre; by 1950 it was about 33 bushels per acre, and colossal

yields of more than 100, and even above 200 bushels per acre, are

reported under exceptionally favorable conditions. Most, though not

Heterosis in Different Organisms 217

all, of this phenomenal improvement is ascribable to the introduction

of the hybrid corn. It is not surprising, then, that by 1950 more than

75 per cent of the corn-producing fields in the United States, estimated

at about 65 million acres, was planted to hybrid corn. In the chief

corn-producing states, such as Iowa, the hybrid corn plantings occupy

close to 100 per cent of the corn acreage. It became urgently neces-

sary to save the original field "varieties" of corn from total extinction.

Their persistence is important because they carry the gene pool from

which new inbred lines can be isolated, which may yield hybrids even

superior to the now available ones.

Heterosis in Different Organisms. The spectacular success of

hybrid corn has inspired attempts to exploit hybrid vigor in other

cultivated plants and domestic animals. These attempts are mostly

in the experimental stage, but at least some of them are yielding en-

couraging results. Heterosis is a phenomenon of obvious practical,

as well as theoretical, importance. The understanding of its biological

nature is, nevertheless, far from satisfactory.

Darwin concluded on the basis of his experiments (see above) that

hybrid vigor arises from the union of diverse heredities serving as a

stimulus inducing powerful growth and general well-being of the

organism. In modern terms, this would mean that heterozygosis for

many genes is per se a viability stimulus. But this cannot be the whole

story. The loss of vigor produced by inbreeding and the luxuriant

development induced by crossing of different strains are much more

pronounced in some organisms than in others. Thus heterosis in corn

is very important, whereas in wheat it is absent or barely detectable.

According to Mangelsdorf, even the best inbred lines of corn yield no

more than half as much as do the open-pollinated varieties from which

the inbred lines are isolated. Many of the inbred lines are so weak

that they can be maintained at all only with difficulty. In contrast,

self-fertilization is the rule in many wheats, oats, barleys, beans, etc.

A field of wheat consists, then, of a single or of several inbred lines

which have been selected because they are vigorous and high-yielding.

In general, the inbreeding depression is most pronounced in species

which normally live in large cross-fertilizing populations, such as corn

or rye. In normally self-fertilizing forms there is little inbreeding

depression and little heterosis upon crossing. Man stands somewhere

in the middle of this heterosis range. History records that some royal

dynasties practiced brother-sister marriage, which is a very close in-

breeding, for many generations without known adverse effects. On

218 Evolution under Domestication

the other hand, recessive hereditary diseases occur most frequently in

families with much consanguinity, that is, where parents are relatives.

Genetic Mechanisms Which Bring about Heterosis. As pointed out

in Chapter 7, "heterosis" is a common name for several different phe-

nomena. Perhaps the simplest of these is mutational heterosis. We

have seen (pages 139-142) that recessive mutants accumulate in popu-

lations of species which reproduce sexually and by cross-fertilization.

No matter how deleterious, or even lethal, may be a mutant when

homozygous, it is "sheltered" in heterozygotes with normal dominant

alleles. A heterozygous carrier of a serious recessive hereditary dis-

ease may enjoy robust health. Thus it happens that in wild popula-

tions of Drosophila, and doubtless of many other organisms as well,

most individuals are carriers of one or more deleterious recessives.

The frequency of any one deleterious mutant in the gene pool, how-

ever, is usually so low that, when individuals who mate are not very

closely related, the probability that both parents would by chance

carry the same deleterious recessive mutants is not very great. The

defective homozygotes are, consequently, rare in the populations so

long as consanguinity is avoided.

Suppose that a recessive gene which is deleterious when homo-

zygous has a frequency q — 0.01 in the gene pool of a population. As-

sume that this population reproduces sexually, that it consists of a

large number of individuals, and that it is panmictic (the matings

occur at random). The frequency of homozygotes who will suffer

from the harmful effects of this gene in this population will, then, be

q3 = 0.012, or one per 10,000 individuals (see Chapter 6, and particu-

larly page 120). The hereditary disease or malformation produced by

the homozygosis for this gene will be a rare one in the population.

But suppose now that we are dealing with a population in which

matings of brothers and sisters occasionally take place. Brothers and

sisters have a fifty-fifty chance of having both inherited a given gene

from their parents. The probability that the siblings who mate will

both carry a recessive which one of their parents carried in heterozy-

gous condition is 0.5 X 0.5 = 0.25, or one-quarter. A quarter of their

progeny is likely to be homozygous for the recessive. Thus one-six-

teenth of the progeny of a brother-sister mating will be homozygous

for any one gene inherited from the grandparents. Self-fertilization

is, of course, a form of inbreeding even more extreme than brother-

sister mating. If a self-fertilizing individual of, for example, wheat

is heterozygous for a recessive mutant gene, a quarter of its progeny

will be homozygous for that gene.

Genetic Mechanisms Which Bring about Heterosis 219

There is no doubt that a part of the degeneration of the progeny

resulting from consanguinity and inbreeding in such species as man

or corn results from homozygosis for recessive deleterious genes. There

is also no doubt that the absence of deleterious effects of consan-

guinity in normally self-fertilizing species is due in part to the rapid

elimination of the deleterious recessive mutants. In wheat a dele-

terious recessive mutant is not sheltered for long in heterozygotes, and

is eliminated by natural selection far more rapidly than it would be

in corn or in man. Nevertheless, mutational heterosis is only a part

of the whole heterosis phenomenon, and at present it is not certain

how large or how small a part.

Another mechanism which causes hybrid vigor is balanced heterosis,

which we have discussed in another connection (Chapter 7, pages

142-147). In Drosophila natural populations of some species consist

mostly of individuals heterozygous for certain chromosomal inversions,

and such individuals possess high fitness. But chromosomal homozy-

gotes are also produced in the populations, and these homozygotes

are less fit than are the heterozygotes. Natural selection maintains a

balanced polymorphism, with a certain optimal proportion of hetero-

and homozygotes arising in each generation. It is certain that bal-

anced polymorphism is widespread in sexual species in which inbreed-

ing is normally rare, but just how widespread we are not certain.

Populations of many species consist of two or more "phases" which

differ in color, in shape of some body parts, or in other traits. Al-

though the phases may appear very strikingly different to the human

eye (Figure 9.8), they interbreed freely both in nature and in experi-

ments. In a number of cases it has been shown (by Ford in some

butterflies, by da Cunha in a species of Drosophila, etc.) that the

phases differ in a single gene or in a few genes, and that their mainte-

nance in nature is due to the heterozygotes being adaptively superior

to the homozygotes. Allison (1954) found that at least one hereditary

disease is maintained in human populations by a similar mechanism.

Heterozygous carriers of the gene for the sickle-cell anemia are more

resistant to malaria than are "normal" homozygotes which do not

carry this gene al all. Yet the sickle-cell homozygotes die of an acute

anemia (see page 141).

Balanced heterosis is likely to be most important in sexual species

which are so well adapted and successful in their environments that

they build large, variable, and panmictic populations. Crow (1948,

1952) came to the conclusion that at least 95 per cent of the hybrid

vigor in corn is due to balanced heterosis, and 5 per cent or less to

Suggestions for Further Reading 221

eral blood groups and of taste blindness is due to balanced polymor-

phism? May the excessively fat and excessively lean persons which

occur in human populations represent the homozygotes, the existence

of which in the species is a necessary by-product of the "normal" well-

proportioned heterozygotes being in a majority in these populations?

Only more research can answer these and other fundamental ques-

tions of basic population biology of man.

Suggestions for Further Reading

Darwin, Ch. 1868. Variations of Plants and Animals under Domestication.

Though the evidence which Darwin had at his disposal concerning the origins

of domesticated animals and plants was much less than is available at present, his

book remains a classic which is well worth reading. It is available in several

editions.

Simpson, G. G. 1951. Horses. Oxford University Press, New York.

A delightfully written as well as scholarly discussion of the evolution of fossil

and living horses.

Hutchinson, J. B., Silow, R. A., and Stephens, S. G. 1947. The Evolution of

Gossypium and the Cultivated Cottons. Oxford University Press, London.

A study of the wild and cultivated cottons and their origins.

Mangelsdorf, P. C. 1947. The origin and evolution of maize. Advances in

Genetics, Volume 1, pages 161-207.

Mangelsdorf, P. C., and Smith, C. E., Jr. 1949. New archeolngical evidence on

evolution in maize. Harvard University Botanical Museum Leaflets, Volume 13,

pages 213-247.

Stebbins, G. L. 1950. Variation and Evolution in Plants. Columbia University

Press, New York.

Chapters VIII and IX contain an excellent discussion of the origin of species by

polyploidy.

Gowen, J. W. (Editor). 1952. Heterosis.

A symposium on heterosis and its applications in agriculture. The articles writ-

ten by J. F. Crow and P. C. Mangelsdorf are particularly interesting.

Lerner, I. M. 1954. Genetic homeostasis. John Wiley, New York.

The most modern discussion of the genetic basis of heterosis and of its evolu-

tionary significance.

/o

Evolution of the Organic Form

and Function

Preformation and Epigenesis. Having retired early from a profitable

business, Antony van Leeuwenhoek of Delft, Holland, became an

amateur microscopist eager to examine anything under his microscope.

In 1675 he examined the seminal fluids of several animals, including

man. He saw swimming in these fluids the "animalcules," or the

spermatozoa as we would say now. This was a discovery enough

to make anybody famous, but a few years later another countryman of

Leeuwenhoek "improved" on it by publishing a picture of a human

spermatozoon in the head of which he saw a "homunculus," a tiny

figure of a man (1694). This seemed a really magnificent discovery,

for it appeared to solve at one stroke the difficult problems of heredity

and development. The human body is all ready, preformed, in the

male sex cell; all it needs to become an adult man is to increase in size.

To be sure, some of the authorities of that time did not feel con-

vinced that the homunculus resides in spermatozoa, and preferred to

look for him in the female sex cell, the egg. But whether they be-

longed to the school of "animalculists" or to that of "ovists," they

believed the idea of preformation to be an excellent one. Especially

so when Jan Swammerdam, of Leyden, Holland, developed the idea

by supposing that the homunculi in the sex cells contain within them

still smaller homunculi; those have more minute homunculi, and so

ad infinitum. The reproductive organs of Adam—or of Eve—contained

within them the entire mankind to come, packed like boxes within

boxes. The utter absurdity of this notion was not obvious at the time,

since it was not realized that within a few generations the homunculi

would have to be smaller than atoms.

The preformation theory, and its extreme version the "box theory,"

222

224 Evolution of the Organic Form and Function

by small steps from inanimate things to animate. . . . After the realm

of lifeless things, there follows the realm of plants. . . . The plants

appear to be animate compared to other things, but inanimate com-

pared to the animals. . . ." Among marine life we find, according to

Aristotle, creatures which occupy steps, intermediate between plants

and animals. Thus sponges are almost exactly like plants, whereas

oysters and mussels are somewhat more animal-like. Then follow

"bloodless" animals (which we now call invertebrates), and those

which have blood (vertebrates). Among the last, we have the differ-

ent steps of perfection represented by fishes, birds, oviparous quadru-

peds (reptiles), and viviparous quadrupeds (mammals). The top

rung of the ladder is occupied by man. Just below man are the

monkeys, which share some human and some quadruped properties.

To Aristotle the ladder of progress did not mean that the occupants

of the upper rungs have arisen by an evolutionary process from those

lower down on the ladder (see Chapter 14). The ladder is just there;

it is preformed. This idea was tremendously popular right down to

the nineteenth century. To Leibniz, the "force of the principle of

continuity" is so great that he ventures to predict that whenever some

rungs of the ladder seem to be unoccupied (that is, when intermedi-

ates between some groups of living beings are unknown), their occu-

pants will be discovered by further studies. This sounds almost like

the assurance of an evolutionist that "missing links" between existing

groups of organisms have existed in the past and become extinct. To

Leibniz it meant only that the preformed order of nature cannot be

incomplete. The Swiss zoologist Bonnet (1720-1793) worked out the

ladder of nature in greatest detail. It begins (starting at the top)

with man—orangutan—monkey—quadruped—flying squirrel—bat—os-

trich—bird—etc. The steps between animals and plants are occupied

by polyps—sea anemones—sensitive plant (Mimosa)—plants; and the

steps between the living-and non-living by truffles—corals—fossils—as-

bestos—talcum. The bottom is "pure earth"—water—air—fire—"finer

matters."

Lamarck (1744-1829), the pioneer of evolutionism, took the leap

into the future by asserting that the more complex organisms have

actually evolved from the simpler ones. The diversity of living

creatures is not preformed, it has arisen in the course of time. But

Lamarck still clung to the notion of a single ladder of progress, al-

though he had it branch in some places. This mistake cost him

dearly. His adversary Cuvier (1768-1833), the great anatomist and

paleontologist, had little difficulty in showing that no single ladder

Types 225

of progress exists in the biological world. The contemporaries thought

that Cuvier had invalidated the idea of evolution as well.

Types. Living creatures are obviously too diverse to be arranged

in a single file from the highest to the lowest. A slight acquaintance

with the anatomy cl a bat will show that, contrary to Bonnet, it is in

no way intermediate between a mammal and a bird. Another notion

can be tried instead—the notion of type, which has already been

mentioned in another connection (page 134). According to Plato

(427-347 B.C.), the world as we see it consists of mere shadows of the

real but invisible world of eternal, unchangeable, and perfect ideas

of things. Men as we see them are variable, and many are miserable

specimens; but there exists in some heaven the Idea of Man of in-

conceivable purity, beauty, and perfection. Animals and plants are

also variable, but an acquaintance with them shows that many or-

ganisms resemble each other, as though they were variations on a

limited number of basic themes. These basic themes were called

types or ground plans.

In 1790 the great poet Goethe (1749-1832) published a biological

work, The Metamorphoses of Plants, in which he devised the "Primeval

Plant" (Urpflanze, in German). According to Goethe, "Everything is

a leaf." What he means is that various organs of the plant, such as

the cotyledons of seeds, petals and sepals of flowers, stamens and

pistils, which are the male and the female organs, are all modifications

of a single structure—the leaf (Figure 10.2). "Through this sim-

plicity," says Goethe, "becomes possible the greatest diversity" of forms

among plants, which he has carefully observed himself and studied in

the works of others. Goethe is quite explicit in saying that the Primeval

Plant which he describes and figures does not actually exist now, and

presumably never existed as an ancestor of all other plants. Although

some historians are eager to make Goethe an evolutionist, he was not

one. The Primeval Plant is the ideal type of a plant, and the plants

that actually grow are varying manifestations of this type. Goethe,

and his younger countryman Oken (1779-1851), tried also to devise a

ground plan of the vertebrate animals, or rather of skulls. The skull

bones are modifications of six vertebrae. Again, this did not mean a

reconstruction of an actual ancestor with six extra vertebrae in place

of a head; it meant only "the concept, the Idea of the animal."

We have already mentioned Cuvier as an opponent of Lamarck. To

Cuvier, all animals were variations of four great types or ground plans:

those of the vertebrates, mollusks, segmented animals (arthropods and

worms), and radially symmetrical animals. Having extensively studied

Homology 227

the fossils found in the vicinity of Paris, Cuvier knew well that his

four types are represented in the different strata by different animals.

Had he studied fossils also in other countries, he might have seen that

the changes of the inhabitants from stratum to stratum meant evolu-

tion. But it so happens that near Paris there are six sharply different

geological formations, the intermediates between which are not rep-

resented. To Cuvier, this meant that life on earth was repeatedly

destroyed by some catastrophes, and repeatedly created anew, and

always according to the same ground plans!

Homology. The idealistic philosophy of Plato, leading to construc-

tion of ideal "types" or "ground plans," proved to be a wrong guide to

the understanding of organic form. Nevertheless, the work of the

typologists yielded many facts that are of enduring value. In com-

paring different animals or plants, we readily notice that some of their

organs and body parts have the same fundamental structure, although

they may be different in appearance and in use. The English anato-

mist Owen proposed in 1843 to call such organs or body parts homolo-

gous. But the phenomenon to which this name applies was known

to Aristotle, who saw quite clearly that human arms are the homologues

of the forelegs of quadruped mammals. In 1555 the Frenchman Belon

compared the skeletons of a bird and a man and identified the homolo-

gous bones, which he indicated by similar letters in the two drawings

in Figure 10.3. Belon's drawings are perfectly acceptable in a manual

of comparative anatomy today.

In modern terms Goethe's Primeval Plant embodies his correct rec-

ognition of the homology between corresponding parts of the same

plant, such as leaves and flower petals (this is called serial homology),

as well as between parts of different plants (special homology).

Cuvier's four types of animal structure were also based on his ability

to perceive the homologies of the corresponding parts in different

animals. The great problem is this: how does homology arise? The

solution of this problem was supplied by Darwin: different organisms

possess homologous organs because they are descended from a com-

mon ancestor. By and large, the greater the similarity in the body

structure, the closer is the common ancestry; the less the similarity,

the more remote is the descent relationship.

Homology does not prove evolution, in the sense that nobody has

actually witnessed the gradual changes in the millions of consecutive

generations which led from a common ancestor to a bird on the one

hand and to man on the other. But homology suggests evolution;

the facts of homology make sense if they are supposed to be due to

230 Evolution of the Organic Form and Function

tion. The student begins to perceive what pieces of the mouth ma-

chinery of a mosquito correspond to the different parts of the cock-

roach mouth. Finally it becomes evident that all the innumerable

insects really have their mouth apparatus built out of homologous

pieces, but these pieces can be vastly different in form. Pre-Darwin-

ian biologists would have said that all insects have their mouth parts

built as variants of the same ground plan. We must add that the only

reasonable interpretation of the generality of the ground plan yet

proposed is that it indicates common descent. All insects have evolved

in the course of time from similar progenitors.

Next, we should ask ourselves this question: what has caused the

mosquito mouth to appear so different from the cockroach mouth?

The reasonable hypothesis is that these insects have different mouth

parts because they feed on different foods. A cockroach bites off

pieces of solid or semi-solid foods with its mandibles, then masticates

the food with its mandibles and maxillae, and finally pushes it down

the mouth opening located at the base of the labium. It has organs

of chemical sense on the palpi of the maxillae and the labium, which

test the quality of the food. The delicate parts of the mouth are pro-

tected by the shield of the labium. A female mosquito gets its food

in an entirely different way, by drawing blood from animals much

larger than itself. With the mouth parts of a cockroach it might

never bite through the skin, or if it did would cause the animal much

pain and a defense reaction. Instead, a mosquito pierces the skin

with the aid of its mandibles and maxillae transformed into finest and

sharpest needles. The operation is often so nearly painless that the

animal frequently does not perceive the presence of the bloodthirstv

insect and does not drive it away. The mosquito sucks the blood up

into its mouth through a fine tube formed by its labium. The mouth

parts of a cockroach would be far less efficient for this purpose, and a

mosquito would be quite unable to feed on the kind of food which a

cockroach eats.

However, we must beware of thinking that the nature of an organ

is explained by finding out the function which this organ performs.

Animals and plants do not get organs just because they need them

or can conveniently use them. To think so would mean ascribing pur-

poses to nature. This is telcological reasoning, which has no place in

science because it explains nothing. But the fear of teleology can be

carried too far. Some biologists go to the extreme of saying that the

function of an organ has nothing to do with its being there. Yet no-

body can deny that man has eyes to see with, and a mosquito has its

Analogy 231

mouth parts to get blood with. It is pedantic to quibble even about

the statement that the purpose of the eyes is seeing. There is really

nothing objectionable about such a statement, which simply describes

what the organ does, provided that one always keeps in mind that

the presence of an organ and its function are at the opposite ends of

a long and complex chain of cause-and-effect relationships. Some of

the connecting links in this causal chain are the processes of mutation,

sexual recombination, and natural selection over a long series of gen-

erations. Darwin has done away with teleology, by giving a rational

explanation of the evolutionary processes which mold an organ so

that it becomes fit for the performance of a given function. (For

further discussion of this problem, see Chapter 14.)

Analogy. Different functions make homologous organs dissimilar

in form. Similar functions may make different, non-homologous or-

gans come to resemble each other. Owen (1843) proposed to call

such organs analogous. A classical example of analogy is the wings

of a bird, a bat, and a butterfly. These organs are unquestionably

similar in function—they are, indeed, wings. But they are very dif-

ferent in organization. A bat wing is a skin fold between the four

fingers of a hand, the bones of the fingers being lengthened to perform

a supporting function, like the ribs of an umbrella. A fifth finger, the

thumb, is short and has a strong claw, by means of which the animal

suspends itself while at rest from a branch of a tree or a ceiling of

a cave. In a bird the supporting surface of the wing is composed of

feathers, not of a skin fold. The skeletal framework of a bird wing

corresponds to that of the whole human arm; but the finger bones

(phalanges) are elongated in only one finger, other fingers being

rudimentary. Now, although the bones of a bat wing and a bird wing

are homologous, the wings themselves are quite differently constructed.

The wings of a butterfly consist of a membrane formed by two layers

of cells, supported by "wing veins" which are thickenings of the mem-

brane. Parts of a butterfly wing are not homologous to parts of either

bat or bird wings. These are analogous organs.

As stated above, early zoologists regarded bats to be intermediate

between mammals and birds (Bonnet, in the eighteenth century; still

earlier Gesner, in the sixteenth, and others). We can see that the

resemblance of bats to birds is due chiefly to analogy, to the presence

of organs of dissimilar structure which, however, serve similar func-

tions—wings for flying. Otherwise the bodies of bats are built like

those of mammals; bats, indeed, are mammals. Whales were consid-

ered gigantic fishes, until John Ray (1627-1705) pointed out that they

232 Evolution of the Organic Form and Function

breathe with lungs instead of with gills, have a heart with two ven-

tricles like mammals instead of a single ventricle like fishes, and

resemble mammals in everything except in the almost complete ab-

sence of hair. And yet whales and porpoises certainly resemble fishes

in body shape more than they do any other mammals. The reason

for this is not far to seek: the streamlined body shape and the presence

of fins instead of legs are greatly advantageous for living in water.

Natural selection has favored in water-dwelling animals, whether in

fish or in mammals, those characteristics which fit them for the water

life. As far as the body shape is concerned, the ancestors of a fish

and a whale were presumably less similar than these animals are today.

The evolution was in this case convergent. But note that the con-

vergence has affected only some traits which have a direct bearing on

the way of life as a water inhabitant. The body structures as a whole

have not converged.

Natural and Artificial Classification. "Without classification—only

chaos," such was the admonition of Linnaeus (1707-1778) to the

young science of biology. Animals and plants must be classified for

the same reason that books must be classified in any large library or

stamps in any large collection. Anybody who has tried to find a book

misplaced in library stacks knows this by experience. But how should

the classification be made? The difficulty arises because different

systems of classification are possible. Stamps are classified usually

by country, then by time of issue, by denomination, and by variations

in the shade of color and of perforation. But books may be classified

according to the name of the author, or by language, or according to

the subject matter, or the date of publication. Living beings vary in

many more ways than either stamps or books; hence the choice of their

classification system is even more complex.

None of the methods of classifying books are intrinsically better

than the others; and each method is more convenient, and is therefore

used in some libraries. The convenience of a classification when it

is being used is, then, the criterion of its validity. The Book of Leviti-

cus divides the water-dwelling animals into those which have fins and

scales and those which do not. This division separates, for example,

the eels from fishes and combines them with a lot of most diverse

invertebrate animals. The Roman Pliny divided the animals into those

living on earth, in water, and in the air. This system combines bats

with birds, whales and porpoises with fishes, and places the am-

phibians athwart the dividing line. Biologists find these classifications

inconvenient because, it is said, they are artificial instead of natural.

Phylogenetic System 233

A natural classification must take into account the greatest possible

number of characteristics of the organisms classified. Thus saying

that an animal is a fish tells a zoologist at once that it respires with

gills, has a heart with a single ventricle, a certain structure of the

brain, of the kidneys, and many other things. A mammal respires with

lungs, has a heart with two ventricles, different brain and kidney struc-

tures, feeds its young with milk, usually has hair, etc. Pliny's classifi-

cation does not do so well for the purposes of a zoologist, because it

tells him at most whether he should look for the animal in water or

out, and next to nothing about the body structure of that animal.

This is, of course, not a valid objection against an artificial classifica-

tion, such as that of the Book of Leviticus, inasmuch as its author had

in mind purposes quite different from those of a zoologist.

In making the natural classification we must, then, distinguish be-

tween homology and analogy. To do this requires a great deal of

study and insight, and it took centuries to erect what we reasonably

believe to be the natural system of the animal and plant kingdoms.

As a matter of fact, Aristotle made a classification which we consider

much superior to Pliny's. For example, he did distinguish mammals,

birds, reptiles, and fishes, although he did not perceive that a cuttle-

fish (a cephalopod mollusk) is an animal really very different from an

ordinary fish. Linnaeus keenly appreciated the advantages of natural

classification, and yet his division of the plant kingdom into twenty-

four classes is pretty artificial by modern standards. The Linnaean

classes were based entirely on the structure of the reproductive organs

(flowers), and fifteen of the classes differ merely in the numbers of

the stamens which a flower possesses. Cuvier did much better classi-

fying animals, but his group Radiata, combining forms as diverse as

corals, sponges, and sea urchins, is an artificial one.

Phylogenetic System. Quite a new understanding of the meaning

of natural classification was given by Darwin. Before Darwin, zoolo-

gists and botanists had to strive for classifications which expressed

the fundamental, rather than only superficial, similarities and dissimi-

larities of the animals and plants. But the classification so arrived at

was completely static; it was just there, preformed in the order of

things. To Darwin the fundamental similarities (homologies) meant

evidence of descent from common ancestors. A classification which

expressed these similarities became, then, a description of the evolu-

tionary development, of the phylogeny, of the organisms so classified.

A natural system is one which puts together the near kin, and sep-

arates the distant relatives.

234 Evolution of the Organic Form and Function

To some of Darwin's successors, especially to Haeckel (1834-1919),

this idea became the cornerstone and the inspiration of their work.

Building the natural system of animals and plants is now more than

construction of a catalogue of organisms for the convenience of fel-

low biologists. It is also a study of the pedigree of the living world,

of the history of its evolutionary development. It is fair to say that

the last decades of the nineteenth century were dominated in biology

by the building of the system of organisms under the guise of study-

ing their phylogeny. The modern classification of animals and plants

was achieved thanks to the labors of a multitude of zoologists and

botanists chiefly of that century.

This immense work may now be regarded as finished in the main,

although, of course, numerous special problems still await solution.

For example, there is no reasonable doubt that mammals, birds, and

reptiles are natural classes. But the Australian monotremes (the

duckbill and the spiny anteater) are a curious group, the evaluation

of which is still in doubt. They have hair and nurse their young like

mammals, and yet lay eggs in reptilian fashion and have some other

odd structural features. Some authorities consider them intermediate

between reptiles and mammals, and related to the ancestral stock by

means of which a group of reptiles gave rise to the mammals. Other

authorities believe the monotremes to be rather an independent off-

shoot from the reptiles, the mammals having arisen from something

quite unlike the monotremes.

In 1894-1896 Haeckel published his great work Systematic Phylog-

eny; a Sketch of a Natural System of Organisms Based on their Descent,

in which he presented what he believed to have been a genealogic tree

of the living world. Many biologists are at present rather less opti-

mistic about the reliability of such genealogies than were biologists in

Haeckel's day. After all, the modern natural classification of organ-

isms is still based on the same kind of data which were used by Lin-

naeus and by Cuvier. To be sure, these data are at present incom-

parably more complete and detailed than they were a century or two

centuries ago, but the phylogenetic interpretation of the natural sys-

tem has not changed the classification to any appreciable extent.

There is no reason to doubt that similarities between organisms usu-

ally indicate common descent, except when the similarities are due to

analogy rather than to homology. However, if we desire to learn

something about the actual history of life on earth, the study of the

now living organisms, no matter how detailed, is no safe substitute

for the study of fossils.

Biogenetic Law 235

Biogenetic Law. The end of the nineteenth and the beginning of

the present century saw most biologists busily constructing phylogenies

of the animal and plant kingdoms. As we have seen, the chief method

used for this purpose was careful comparative study of the body struc-

tures of diverse organisms, which usually permitted homologies to be

distinguished from analogies. Another and almost equally potent

method was the study of the development of the organisms, especially

of the embryonic development.

The bodies of animals and plants are often most intricately and

wonderfully built. Yet these bodies arise from very obscure begin-

nings, single egg cells fertilized by single sperms, and by means of

developmental processes which have seemed utterly mysterious. The

preformation and box theories (see above) tried to solve the problem

by pushing it back to the act of original creation, but this proved a

wrong lead. Examination of the embryos of different organisms

showed at once that these theories lacked foundation. Nehemiah

Grew (1682) found that an apple seed contains a structure which in

no way resembles a miniature apple tree. Animal embryos were stud-

ied beginning with Fabricius of Aquapendente in Padua (1537-1619)

and his great English pupil, William Harvey (1578-1657). Malpighi

(1628-1694) was apparently the first to apply to this study the then

novel instrument—the microscope. Wolff, von Baer (see page 223),

Purkinje (1787-1869), and their successors built the science of em-

bryology, which progressively unraveled the developmental stages of

all classes of living beings. Everywhere the development proved to

be epigenetic rather than preformistic.

Embryologists were impressed by a rather different aspect of the

matter. In 1821 Meckel concluded: "Embryos of higher animals pass,

before they complete their development, through a succession of stages.

. . . The embryos of higher animals, of mammals and especially of

man, pass through stages resembling more or less completely, in the

appearance of separate organs as well as of the whole body . . . the

lower animals." Indeed, the body starts by being a single cell, and is

to that extent like the unicellular organisms. Then it becomes com-

posed of two layers of cells, the ectoderm and the endoderm, like a

hydra or a coral. Especially impressive was the discovery that mam-

malian and human embryos at a certain stage have in the neck region

gill pouches and gill bars. In these embryos these structures never

function in respiration, but in the fishes they do. A mammalian em-

bryo, then, passes a "fish stage." The embryonic development of

higher organisms seems to retrace a sequence of body structures cor-

236 Evolution of the Organic Form and Function

responding first to the lower and then to progressively higher forms.

Darwin pointed out that the above observations are most easily

understood if the higher organisms are the evolutionary descendants

of the lower ones. Haeckel, who had a talent for making pithy for-

mulae, expressed the idea thus: "Ontogeny (the development of an

individual) is a brief and rapid recapitulation of the phylogeny," and

he gave to this formula the ambitious name of the "basic biogenetic

law" (1866). If the embryo of an organism is an archive which con-

rains a telescoped evolutionary history, the study of the embryonic

development becomes a method of unraveling the phylogeny of the

group. Haeckel and his many followers eagerly proceeded to do just

that. The result was accumulation of accurate and detailed data on

the embryology of all kinds of organisms. However, most biologists

are at present inclined to believe that the "basic biogenetic law" rather

overstated the extent to which the embryonic development is an ac-

curate record of the past evolutionary history. The situation may be

understood best by considering two examples of the evidence rele-

vant to the issue.

Development of the Urogenital Organs in Vertebrates. Three kinds

of kidneys occur among the vertebrates, known as pronephros, meso-

nephros, and metanephros. A pronephros is a series of tubules, which

draw waste products not from the blood stream but directly from the

liquid filling the spaces between the organs in the body cavity. These

waste products are conducted to the outside of the body by means of

a pair of tubes—the pronephric ducts. The pronephros functions as a

kidney only in the most primitive vertebrates, such as young lampreys

(Cyclostomata); but a pronephros is nevertheless formed in the em-

bryos of all other vertebrates. In the mammalian embryo it never

functions as an organ of excretion, but it is there just the same (Figure

10.5).

Next, there appears in the embryos of all vertebrates a pair of kid-

neys of quite different structure—the mesonephros. These kidneys

have an abundant supply of blood vessels from which the renal tubules

draw the waste products. The mesonephros communicates with the

outside by means of a duct called the Wolffian duct, which develops

by modification of the pronephric duct. This is the functional kidney

of most fishes and of amphibians, but it is formed, next to the proneph-

ros, also in the embryos of the reptiles, birds, and mammals. In these

embryos, however, it is only a temporary arrangement, which soon

disappears, leaving behind the Wolffian duct which is put to other

uses (Figure 10.5).

238 Evolution of the Organic Form and Function

In the higher vertebrates there arises a third kidney, the metaneph-

ros, which is the urine secretor in the adult animals. It has a duct

of its own, called the ureter. The metanephros and its ureters are

wholly absent in fishes and amphibians.

The development of the excretory organs, the kidneys, is intimately

connected with the development of the reproductive organs. In the

frog and other amphibians the sperm produced by the testes is con-

ducted to the outside through the mesonephric, Wolffian, ducts. The

eggs from the ovary are picked up by another pair of ducts, the

Miillerian ducts. These two pairs of ducts, the Wolffian and the

Miillerian, are formed in the embryos of most vertebrates. Thus the

human embryo about six weeks old has both of them. In the male

embryo the Wolffian ducts persist and become the sperm ducts, losing

all relation to the adult kidneys in the reptiles, birds, and mammals.

The Miillerian ducts disappear, except for small vestiges. In the

female, the Miillerian ducts develop into the oviducts, and in mammals

also into the uterus and a part of the vagina. The Wolffian ducts dis-

appear, leaving behind two rudiments in the ovary (Figure 10.5).

Why should pronephros and mesonephros, which are the kidneys of

the lower vertebrates, be formed and then disappear in the embryos

of the higher vertebrates? This would certainly be inexplicable if

the higher vertebrates were not the descendants of the lower. Again,

an anti-evolutionist would have to blame God for having arranged

things just to mislead the zoologist. The biogenetic law does contain

an important kernel of truth. But it does not follow that a human

embryo is in a lamprey stage when it has the pronephros, and in a

frog stage when it has a mesonephros. At these or at any other stages

the human embryo has no far-reaching resemblance to any adult ani-

mal. Von Baer stated the facts clearly as early as 1828, when he wrote:

"An embryo of a higher animal is never basically like any other ani-

mal, only like the embryo of the latter." In other words, the develop-

ment of the embryo does not reproduce for our benefit the portraits

of the near or remote ancestors. But the evolutionary descent is mani-

fest in embryos forming certain organs homologous to those found in

the adults of quite different, but phylogenetically related, organisms.

Larvae of Some Crustaceans. Just as with the embryos, remark-

able similarities are often found between the larvae of animals which

are quite dissimilar as adults. Many lower crustaceans, such as the

little copepod Cyclops, produce eggs from which hatches a very tiny

but characteristic free-swimming larva, called the nauplius (Figure

10.6). The nauplius, with its three pairs of legs, eventually meta-

240 Evolution of the Organic Form and Function

morphoses into an adult Cyclops, having several pairs of legs and leg-

like appendages, a complex nerve system and sense organs, as well

as organs of reproduction. The barnacles look in the adult condition

about as unlike a Cyclops as can be imagined. Instead of swimming

freely in water (in the plankton), a barnacle is attached to rocks or to

the surface of other organisms on the sea bottom or in the tidal zone.

It usually has a hard shell consisting of several calcareous pieces, in-

side of which is found an animal glued to its support by its head.

The classification of barnacles was long in doubt. Even so perspica-

cious a zoologist as Lamarck placed them in his phylogenetic system

between the segmented worms and the mollusks. It was not until it

was discovered that the barnacles begin their existence as typical

nauplius larvae that they were placed as aq order, Cirripedia, among

the crustaceans (Figure 10.6). Some other close relatives of the Cy-

clops and of the barnacles are in the adult condition unlike either.

Thus the Sacculina is a parasite living on crabs. Its body is a shapeless

sack containing chiefly the enormous reproductive organs; this sack is

connected to a system of root-like processes, which penetrate inside the

crab's body and envelop the digestive tube of the latter. Sacculina

needs no digestive organs of its own because it absorbs the food di-

gested by the gut of the crab! And yet the larval form of this para-

site which has lost almost all semblance of animal form is still a free-

swimming nauplius (Figure 10.6).

Taking the biogenetic law literally, we would have to conclude that

the ancestor of Cyclops, of barnacles, and of Sacculina was a kind of

nauplius which reached sexual maturity and reproduced in a state

which in its descendants is a passing larval phase. But no such animal

actually exists, and it is regarded unlikely that it ever existed. The

nauplius phase is there not for the enlightenment of zoologists but

because having it is useful to the creatures. Barnacles and Sacculina

do not move freely when mature, the nauplius larva being the stage

in which they become dispersed and find places to live. The develop-

mental and embryonic stages, like the adult condition, are subject to

evolutionary changes. Mutations occur which alter the embryo, and

natural selection perpetuates the genotypes which yield developmen-

tal patterns most advantageous not only in the adult stage but also

during the whole lifetime.

The common ancestor of the Cyclops and of the Cirripedia was not

a nauplius, but it probably did pass a nauplius-like larval stage. Hav-

ing this kind of larva proved selectively advantageous for many of its

descendants, although the adult forms became adapted to quite dif

242 Evolution of the Organic Form and Function

Why should man possess a token of a tail and a whale have a trace

of the pelvis? Why should mammalian embryos develop pronephros

kidneys which they never use? There was a time when the zoologist

Severino (1586-1656) and the botanist de Candolle (1778-1841) re-

garded vestigial organs simply as souvenirs left by nature to indicate

the type, or ground plan, which it followed in constructing a given

creature. With Darwin and his followers the vestigial organs became

the show pieces of evolutionism. Man's coccyx testifies that our

ancestors possessed a waggable tail; whale's pelvis is there because

whale's ancestors had four legs and made use of their pelves. Human

embryos have a pronephros because our fish-like ancestors had one.

There is, indeed, no doubt that vestigial rudimentary organs silently

proclaim the fact of evolution. But evolution does not explain their

presence completely. After all, some organs may be lost in evolution

without leaving a trace. The vermiform appendix of the caecum in-

testine is a vestigial organ in man, but the intestine of the cat has no

vermiform appendix. If the human coccyx, the vermiform appendix,

or the breasts in male mammals are really useless to their possessors,

why did they not disappear entirely?

To understand this, we must remember that heredity, development,

and evolution are essentially epigenetic and not preformistic. We do

not inherit from our ancestors, close or remote, separate characters or

organs, functional or vestigial. What we do inherit is, instead, genes

which determine the pattern of developmental processes. The fer-

tilized egg is a single cell which becomes many cells; these cells

become compounded into various organs and acquire various physio-

logical functions; the body grows, reaches a stage when it is capable

of reproducing its like, and finally becomes old and dies. Embryonic

recapitulations and vestigial organs are integral parts of the develop-

mental patterns which the now living organisms have inherited from

their evolutionary ancestry. The coccyx arises in the process of build-

ing the spinal column; the development of a whale embryo follows

some of the paths which are common to all mammals and involve

formation of a pelvis and of hind legs; the pronephros appears because

the urogenital system arises by this method. When we talk or write

about the anatomy of man or of some other organism, we are likely

to represent the body as the sum of organs or tissues which develop

independently in ontogeny and in phylogeny. But in reality it is the

developmental system as a whole which is preserved or modified by

natural selection in the process of evolution. Haeckel's biogenetic

Biochemical Homology 243

law, in the last analysis, is simply an expression of the unity and of a

relative stability of the developmental system in evolution.

Biochemical Homology. Morphology, the study of the external and

anatomical structure of the body, developed in the history of biological

science before physiology, which investigates the processes which take

place in the organisms. The theory of evolution was a great generali-

zation derived chiefly from morphological evidence. We have, how-

ever, pointed out that a body structure is always an outcome of de-

velopmental processes which are basically physical and chemical. A

"trait," such as color or shape of some body part, is an outward sign of

physiological processes which have brought it about. The current

century has seen rapid progress in physiology, and it is now becoming

possible to trace physiological evolution just as classical evolutionists

traced its morphological aspect. Some biologists even believe that

"our final theory of evolution will see it largely as a biochemical proc-

ess" (Haldane, 1937).

Almost two centuries ago Lavoisier, the father of chemistry, discov-

ered that the process of respiration resembles in a general way that of

combustion. A molecule of sugar combines with six molecules of oxy-

gen to give carbon dioxide, water, and heat, thus.

C6H12O6 + 6O2 -» 6CO2 + 6H2O + Energy

But the work of the last quarter of a century has shown that the

combustion in the living body is a vastly more intricate and interesting

process, which occurs with the aid of many enzymes and involves

many intermediate steps. A description of this process, which has

been studied in most detail in the muscles of higher animals, can be

found in any modern textbook of physiology or biochemistry. In a

most general way, complex sugars (such as glycogens in animals or

starches in plants) are first split to the simplest sugar, glucose. Glu-

cose is combined with phosphoric acid ("phosphorylated") to a glucose

phosphate, and, through some ten further reactions facilitated by a

series of specific enzymes, turned into pyruvic acid (C3H4O3) The

phosphorylation occurs with the aid of a remarkable substance, adeno-

sine triphosphate, abbreviated ATP. ATP has a high energy content,

which it transfers to the phosphorylatod glucose. The degradation of

the glucose phosphate to pyruvic acid involves, however, liberation of

much energy which is used up in the functioning of the living cell

(such as contraction of the muscle). Pyruvic acid is then broken

down to carbon dioxide and water by an even more complex system

of enzymatic reactions, known as the "citric acid cycle." The im-

244 Evolution of the Organic Form and Function

portant thing here is that the chemical reactions are reversible, so that

the processes can go in either direction, depending on the presence

in the cells of various substances involved and on the need for or the

availability of excess energy. Such an arrangement, according to the

principles of physical chemistry which we need not consider here, is

most economical in that it does not waste energy on heat production,

and in that the living cell can respond adaptively to changes in its

environment.

It is most significant for our purposes that this elaborate chemical

machinery is remarkably widespread in the living world. It is found

not only in quite different kinds of cells in higher animals (muscles,

liver, kidney, nerve cells), but also in quite different organisms, such

as insects, mollusks, protozoans, plants, yeasts, and even bacteria. It

is at present impossible to tell how great may be the differences

between the processes of cellular respiration in different groups of

organisms. But it is very clear that these processes have much more

in common in most diverse organisms than any biologist would have

ventured to speculate until these similarities were actually found. We

are dealing here with most striking biochemical homologies which

attest that the whole living world is really one large family adapted

to subsist in different manners in different environments. An anti-

evolutionist would have a hard time to account for the inexplicable

caprice of a deity who chose to install exactly the same enzymatic

mechanisms in a human body cell and in a yeast cell.

Biochemical Variations and Evolutionary Relationships. Here we

may consider just one example of the kind of biochemical evidence

which throws light on the evolutionary relationships of morphologically

quite different organisms.

The evolutionary derivation of the phylum of vertebrates, to which

belong the higher animals from fish to the gentle reader of this book,

has been a happy hunting ground for speculation among zoologists.

Every major animal phylum was at one time or another considered a

possible ancestor or cousin of the vertebrates; there was even an at-

tempt to derive the vertebrates from fossil relatives of modern spiders,

the so-called water scorpions. The least strained of these speculations

related the vertebrates to the echinoderm phylum (sea urchins, star-

fishes, brittle stars, sea cucumbers, etc.). The chief evidence in favor

of this view is an obscure marine animal, the acorn worm (Balano-

glossus, Figure 10.8). This animal has its nerve system located on the

dorsal side from the gut, as in all vertebrates, and not on the belly

(ventral) side, as in the invertebrates. It has gill slits like a fish

246 Evolution of the Organic Form and Function

creatine phosphate, like the vertebrates. At least one species of star-

fish has only arginine phosphate. The only other animal which is

unorthodox in this respect is again the acorn worm. Some species of

Balanoglossus have only creatine phosphate, but one species has been

reported to have both creatine and arginine phosphates.

This does not mean, of course, that the vertebrate animals have

descended from sea urchins or from acorn worms, any more than

Darwin's theory meant that "man is descended from a monkey." The

now living echinoderms are about as unlikely ancestors of the verte-

brates as any animal can be. The biochemical machinery of the echino-

derm muscles shows, however, a spectacular variability, such as could

be expected in a group of animals the ancestors of which were also

the ancestors of the acorn worms and of the vertebrates.

Serological Reactions. One of the most remarkable adaptive re-

actions in the higher animals is their ability to protect themselves

against the invasions of pathogenic microorganisms by formation of

antibodies (see page 13). When bacteria or viruses, alive or killed,

are introduced into the blood stream, there appear in the blood serum

(the liquid portion of the blood) chemical substances, antibodies,

which can kill or neutralize the microorganisms of the same kind which

induced the formation of the antibodies. This natural protective

reaction, however, may be used experimentally to produce data bear-

ing on the problems of biochemical similarities and dissimilarities

between organisms.

Suppose that human blood, or minced tissue, is injected into a suit-

able animal, such as a rabbit. After a while the serum of the injected

rabbit will contain antibodies to the human blood. This change can

be detected by mixing in a test tube some serum of the immunized

rabbit with human serum; the "immune reaction" will be manifested

by the formation of a precipitate which will settle on the bottom of

the test tube. Of course this immune reaction will develop with a

serum of any human being, not only that of the individual who fur-

nished the material injected into the rabbit. This fact means that all

human beings are chemically similar enough so that a rabbit im-

munized against the serum of one is immunized to all human sera.

Next, the serum of a rabbit immunized against human serum may

be mixed with the serum of a chimpanzee or other anthropoid apes. A

precipitate is formed, showing that the chimpanzee blood proteins are

much like the human ones. But the serum of a monkey, such as a

baboon, gives a very much weaker immune reaction with less pre-

cipitate. Here the chemical similarities are evidently more limited

Gene Homology 247

but still perceptible. A dog serum gives little or no precipitate at all.

Man and dog sera are chemically very different.

With proper technical refinements the serological tests are made to

give quantitative data which may be used to evaluate the biochemical

(or, more precisely, serological) resemblances and differences between

organisms. Such serological comparisons have been made for a va-

riety of animals and also for some plants. By and large, the serological

differences go hand in hand with the differences in the body structure,

and consequently with the position of the organisms involved in the

modern natural classification system. This being so, some interesting

attempts were made to use serological reactions as a criterion of rela-

tionship where the usual morphological data yield not wholly satis-

factory results. Thus some zoologists consider rabbits and hares to

be so different from mice, squirrels, guinea pigs, and beavers that they

put the rabbits and hares in a separate order, Lagomorpha, and the

latter group in the old order Rodentia (rodents). Other zoologists

consider the lagomorphs and rodents to be members of a single order,

Rodentia. Moody and his collaborators found that rabbits are sero-

logically very unlike the rodents. In fact, their data show a closer

similarity between a rabbit and cattle than between a rabbit and a

guinea pig or a rat. This is, of course, evidence in favor of giving the

lagomorphs the status of a separate order.

Gene Homology. The resemblance between the hearts of a whale

and of other mammals is unquestionably due to homology, whereas

the resemblance between the fins of a whale and of a fish is due to

analogy. The whale is undoubtedly a mammal and not a fish, but the

distinction between homology and analogy is not always so easy. The

lagomorphs and the rodents have very similar gnawing incisor teeth.

Should we give a greater weight to the morphological similarity of the

teeth, or to the biochemical difference revealed by the serological

tests? Have the lagomorphs and the rodents inherited their gnawing

incisors from common ancestors or acquired them independently by

convergent evolution? Because of the possibility of such doubts, it is

interesting that the phenomenon of homology extends right down to

the gene level of biological organization.

Suppose that half of the male progeny of a Drosophila female are

red-eyed and the other half are white-eyed (cf. page 61). Ignoring

the possibility of mutation, we can say that the red-eyed males carry

the same gene for red eyes, or, more precisely, the division products

of the red gene which their mother carried in its X-chromosome. Simi-

larly, the white-eyed sons have white genes descended from the white

248 Evolution of the Organic Form and Function

gene in the other X-chromosome of their mother. We can take a long

further step and say that the red and the white genes, which act as

alleles, are both descendants of a gene which was present in a common

ancestor of the flies crossed. Indeed, mutations which produce white

alleles from red have been repeatedly observed. The white and the

red alleles of the "white gene" in Drosophila melanogaster are certainly

homologous.

Drosophila simulans, a species very similar to Drosophila melanogas-

ter, also has red eyes, and also produces by mutation white-eyed strains.

Does Drosophila simulans have a homologue of the "white gene" of

Drosophila melanogaster? We can be sure that it does, because the

two species can be crossed; and, although the offspring are sterile,

they show that the eye color in the hybrids between the species is

inherited exactly as it is within either parental species. Sturtevant

has found by means of similar evidence that Drosophila melanogaster

and Drosophila simulans carry a number of homologous genes. In

fact, no gene has been found present in one species but lacking a

homologue in the other. Gene homology was demonstrated by similar

experiments also in some other pairs of species that can be crossed

and produce hybrids. Harland, Stephens, and others have carried out

particularly elegant investigations of this sort in several related species

of cottons (concerning these species, see Chapter 9). The presence

of homologous genes in different species is, of course, the strongest

possible evidence of the descent of these species from a common an-

cestor, short of actual resynthesis of these species in experiments (cf.

Chapter 9).

Many species cannot be crossed, or produce hybrids which fail to

survive to the stage when certain body traits, such as coloration, de-

velop. Such species may nevertheless produce strikingly similar mu-

tants. For example, white-eyed mutants due to sex-linked recessive

genes are known in about a dozen different species of Drosophila.

Even more remarkable is that different species of rodents and lago-

morphs (domestic mouse, rat, guinea pig, rabbit) and even repre-

sentatives of other orders of mammals (cattle, sheep, horse, dog, cat,

man) give rise to kindred mutants. Thus most of the species named

have albino forms due to recessive gene alleles, genes which produce

characteristic kinds of spotting, genes which give the "wild" (agouti)

coloration of the pelage, etc. A not very rare dominant gene in man

gives a light "forelock" of hair, as well as an albinotic condition of parts

of the skin on the face, chest, and abdomen (the manifestation of this

gene is particularly striking in normally dark-skinned races). It is

Are Genes Preformistic? 249

tempting to compare this gene with genes which give a similar dis-

tribution of white parts in other animals, such as dogs and cats.

Different species, genera, and even different orders of mammals may

still have some homologous genes, descended from the genes which

were carried in their common ancestors. This is, indeed, most likely

justified, at least for the parallel color mutants in the different rodents,

but we must beware of placing too much reliance on inferences of this

kind. When the gene homology in different species cannot be dem-

onstrated by crossing these species and examining their hybrids, the

phenotypically similar mutants may be analogous rather than homolo-

gous. That this danger is a very real one is shown by the existence

within the same species of phenotypically similar mutants which are

produced by non homologous, and non-allelic, genes. For example,

several eye-color mutants are known in Drosophila melanogaster which

differ from the normal fly by having a bright red eye. It is hard to

distinguish these mutants (called vermilion, cinnabar, scarlet, cardinal,

etc.) by inspection, and yet the genes which produce them lie in dif-

ferent chromosomes, and the hybrid offspring of the different recessive

mutants have a normal eye color. Human genetics has also several

instances when similar traits are produced by different genes. For

example, there are at least two different genes which give color blind-

ness.

Are Genes Preformistic? The discovery that many fundamental

biochemical processes are similar in most diverse organisms (see

above) makes the problem of gene homology especially difficult. It

is the more so since the important work of Beadle and his school has

developed the idea that an intimate relationship exists between genes

and enzymes. Working particularly with the fungus Neurospora,

this school found, especially after 1940, that many mutations induced

by ultraviolet treatments and other means have each just one step

blocked in some metabolic reaction chain. It looks as though every

gene is responsible for the production of just one enzyme which

mediates some chemical reaction in the metabolic machinery of the

cell, or at least that there is a "one gene-one function" relationship in

the cell physiology.

Several enzymes performing apparently identical functions have

been discovered in organisms as different as mammals and yeast or

bacterial cells, and there is every likelihood that such discoveries will

be multiplied in the years to come. Should we conclude on this basis

that man and bacteria still possess some genes in common? Some

biologists regard such a conclusion as likely and others as improbable.

250 Evolution of the Organic Form and Function

The attitudes of different biologists towards this unsolved problem

reflect in a most interesting manner the preformist and the epigenetic

types of thinking which, as we have seen, are traceable in biological

science almost from its inception.

As described in Chapter 4, the early ideas concerning the nature of

heredity were clearly preformistic. Darwin's hypothesis of pangenesis

assumed a two-way relationship between the organs of the body and

the sex cells. The organs and tissues manufacture each their own

vestiges called gemmules; the gemmules are transported to the re-

productive glands through the blood stream or similar means; in the

developing embryo each gemmule gives the organ of which it is

the likeness. According to Weismann, whose views were influential

around the turn of the century when genetics got its start, there ex-

isted a strictly one-way street between the sex cells and the body.

The germ plasm in the sex cells consists of "determinants" which re-

produce themselves by division. The development is essentially a

sorting out the determinants present in the fertilized egg, until each

organ and cell comes to contain only the determinant which makes

that cell what it should be in the finished organism. Surely Weismann

has not imagined a "homunculus" hidden in a sex cell or in a chromo-

some, but in effect his determinants amounted to a sort of a dismem-

bered mosaic the stones of which could make a homunculus when

put in the right order. And, incidentally, it was when Weismann

attempted to figure out how this mosaic was put together in the em-

bryonic development that his ingenious theory broke down.

The gene of the geneticists was, to begin with, uncommitted either

to preformism or to epigenesis. Johannsen, who invented the word

"gene" (1909), was especially anxious to keep it austerely pure of

any such implications; to him the gene was just "something" in the

sex cells which determines various "properties" of the organism. He

wisely refrained from speculating about how the determination oc-

curred. But the purity did not last long, partly because of the lan-

guage which biologists were using. When one speaks of the gene

for white eyes in Drosophila, a wrong impression is created that this

gene is concerned with the development of the eye color and nothing

else (see, however, the discussion of manifold effects of genes, page

35). The preformism put out by the front door returned by the back

door. Every gene was named for something, an organ or a character,

or a disease. A visible character, however, is a sign that some physio-

logical reaction had taken place. When physiologists and biochemists

Suggestions for Further Reading 251

began to decipher the chemical nature of these reactions, the gene

became tied to an enzyme or to a "function."

There the matter rests for the present. More knowledge about the

nature of the gene action and of the embryonic development is needed

to throw new light on the fundamental biological problems involved

here. In conclusion, however, it should be stressed that similarity of

the known end effects of genes does not prove that these genes are

identical. Consider, for example, the fact that all classes of vertebrates

have eyes of basically the same structure. Harland pointed out

(1936) that it does not follow from this that the same genes make the

eyes of a fish, of a bird, and of man. In fact, there may not be any

particular genes which make the eye in any of these. The formation

of the eye in the development is a part of the development of the

embryo as a whole, which is governed not by any one gene but by

the whole genotype. Of course changes (mutations) of some genes

are known to strike at some parts of the developmental machinery

(such as the development of the eye or of some of its parts) more than

at other developmental processes. However, it is not only possible

but, indeed, likely that different, non-homologous and non-allelic, genes

may show such effects in different organisms. Muller (1939) has

called this the change of the gene function: the role which a gene

plays in the development of the organism does not remain constant

in the evolutionary process, even though some organs (such as eyes)

remain clearly homologous for long times during the evolutionary

history. The presence of the same enzymes in human beings and in

yeast cells means, then, that these enzymes play adaptively very im-

portant, and perhaps irreplaceable, functions in the cellular metabo-

lism. But even if the specificity of each enzyme is impressed upon it

by one and only one gene, it is probable that different genes do it in

man and in the yeast.

Suggestions for Further Reading

Locy, W. A. 1925. Growth of Biology. Holt, New York.

Nordenskiold, E. 1928. The History of Biology. Knopf, New York.

Darlington, C. D. 1953. The Facts of Life. Allen & Unwin, London.

Darlington's book contains a very interestingly written, if partisan, account of

the history of some basic ideas of genetics and evolution. The other two books

furnish good factual descriptions, but their interpretations are not always in tune

with modern biology.

Buchsbaum, R. 1948. Animals without Backbones. 2nd Edition. Chicago

University Press, Chicago.

252 Evolution of the Organic Form and Function

De Beer, G. R. 1951. Embryos and Ancestors. 2nd Edition. Oxford Univer-

sity Press, New York.

Gregory, W. K. 1951. Evolution Emerging. Macrnillan, New York.

Romer, A. S. 1949. The Vertebrate Body. Saunders, Philadelphia.

Simpson, G. G. 1944. The principles of classification and a classification of mam-

mals. Bulletin of the American Museum of Natural History, Volume 85.

Young, J. Z. 1950. The Life of Vertebrates. Clarendon Press, Oxford.

These six books (Buchsbaum to Young) furnish excellent accounts of compara-

tive anatomy and embryology viewed in evolutionary perspective.

Baldwin, E. 1948. An Introduction to Comparative Biochemistry. Cambridge

University Press, New York.

Florkin, M., and Morgulis, S. 1949. Biochemical Evolution. Academic Press,

New York.

Haldane, J. B. S. 1954. The Biochemistry of Genetics. Allen & Unwin, London.

Wald, G. 1946. The chemical evolution of vision. The Harvey Lectures, Series

41, pages 117-160.

Wald, G. 1952. Biochemical evolution. In Modern Trends in Physiology and

Biochemistry. Academic Press, New York.

Physiology and biochemistry are the biological disciplines which have thus far

been least influenced by evolutionary thought. The works cited above represent

pioneering attempts in studies on the evolution of biochemical systems.

Evolution of Sex

According to Plato's romantic myth, the world was at some remote

time populated by perfect beings who were female on one side of the

body and male on the other. Then angry gods sundered the two

sides, and made the detached halves, females and males, forever seek

to restore the lost wholeness in love. Until the beginning of the cur-

rent century, Plato's myth was about as good an elucidation of the

origin and meaning of sex as could be had. Aristotle, Plato's more

realistic-minded disciple and rival, declared that the female supplies

the matter and the male the motion of the future life. He thought

that the female was "cold" and the male "hot." Consequently, the

conceptions which occur when warm winds are blowing give more

males, and cold winds bring more females. The conjectures and spec-

ulations concerning sex which were being solemnly discussed in Dar-

win's day were not much above the level reached by Aristotle more

than two thousand years earlier.

Two discoveries changed the situation at the turn of the twentieth

century. First, the microscope revealed the existence of sex chromo-

somes (see Chapter 3). In organisms with separate sexes every cell

of a female body differs in the chromosome complement from every

male cell. The behavior of the sex chromosomes at meiosis explains

very- simply how the sex of an individual is decided at the moment

when the egg is fertilized. Second, an insight into the biological mean-

ing of sex was obtained by deduction from Mendel's discoveries.

Mendel's work showed that gene recombination in sexually produced

progenies creates an immense amount of genetic variability and of

raw materials for evolution (Chapter 2). Sex arose in organic evo-

lution as a master adaptation which makes all other evolutionary

adaptations more readily accessible.

253

Female and Male Sex Cells 255

period of rest, or it may undergo at once two meiotic divisions. These

divisions yield four cells, each with half as many chromosomes as were

present in the zygote; the cells separate and start a new cycle of inde-

pendent existence and asexual reproduction.

The critical events in any sexual process are two, fertilization and

meiosis. Both are observed in the alga shown in Figure 11.1, but it

may be noted that the cells which fuse are so much alike in appearance

that neither of them can be regarded as female or as male. The two

sexes in these primitive organisms are called simply plus and minus.

That the plus sex is genetically different from the minus can be dem-

onstrated by a simple experiment. Single cells are isolated, and their

asexual progenies (clones) are kept in separate cultures. No sexual

fusions take place within a clone, since all cells are of the same sex.

But when cells from different clones are mixed, pairs may be formed.

Clones which give rise to pairs are of different sex. The difference

between the plus and minus sexes is inherited very simply; among the

four cells which arise from a zygote as a result of meiosis, two cells

give rise to plus and two to minus clones.

A situation, known as relative sexuality, which at first sight seems to

be paradoxical, arises in some algae and some lower fungi which have

not two but several "sexes." Figure 11.2 illustrates a case when eight

"sexes" can be distinguished, beginning with "strong plus" (+4),

"weak plus" (+1), "weak minus" ( — 1), and so on to "strong minus"

(—4). Any plus strain gives sexual fusions with any minus strain;

but a strong plus gives fusions also with a weak plus, and a strong

minus with a weak minus. The extremes of each sexual group behave

towards each other as representatives of different sexes. The inheri-

tance of relative sexuality is rather complex and not completely worked

out.

Female and Male Sex Cells. In unicellular organisms every cell

may under proper environmental conditions act as a gamete, that is,

participate in a sexual fusion. In multicellular creatures there is a

division of labor between cells of different tissues; cells are specialized

to perform different functions. An animal body consists of skin cells,

muscle cells, bone cells, glandular cells, nerve cells, and finally sex

cells which carry the function of reproduction. The sex cells which

unite may be alike or they may be different. In some marine brown

algae the sex cells are as much alike as the gametes shown in Figure

11.1. Here we cannot, therefore, speak of female and male sex cells.

But in other brown algae one of the gametes that fuse is much larger

than its partner. The larger gamete contains a supply of nutritive

256

Evolution of Sex

materials; it is the female cell. The smaller gamete contains little

food supply but it is efficiently motile; it is the male gamete.

The differences between female and male sex cells are the outcome

Sex

?4

?3

?2

?i

£i

c/2

£3

£4

?

4

—

—

•eak

fncuRim

strong

strong

strong

strong

?

3

-

—

—

-

strong

strong

strong

sbOK

?

2

•ok

—

-

—

strong

strong

stnxig

strong

?

1

medium

•Mk

—

—

«*.

strong

strong

strong

4

i

strong

strong

strong

medium

—

—

-

medium

/

2

strong

strong

strong

strong

Hermaphroditism

257

Hermaphroditism. However striking are the differences between

female and male gametes, eggs and spermatozoa, both kinds of sex

cells mav be produced in the body

of the same individuals. An organ-

ism which gives rise to female as

well as to male sex cells is called a

hermaphrodite. Many invertebrate

animals are hermaphrodites. A ma-

jority of flowering plants are also

monoecious or hermaphroditic. A

flower usually contains female parts

which produce ovules, and male

parts which give rise to pollen

grains. Some plants, such as wil-

lows, however, are dioecious or bi-

sexual. A willow tree produces

either female or male flowers but

not both.

In snails and slugs (pulmonate

mollusks), every individual has a

single hermaphrodite gland which

produces eggs and spermatozoa at

the same time. The sexual ducts

which carry these eggs and sperma-

tozoa to the outside, however, are

separate, and so are the female and

male copulatory organs (Figure

11.3). Self-fertilization of the eggs

by the spermatozoa of the same in-

dividual occurs only rarely. The

usual situation is that two hermaph-

roditic snails mate, and the eggs of

each are fertilized by the sperma-

tozoa of the other. In earthworms

the male and the female sex glands

are separate, and so are the sexual

ducts. The copulation leads usu-

ally to mutual insemination. Her-

maphroditism is rare among insects

and among vertebrate animals.

Among the latter it occurs only in

Figure 11.3. Reproductive organs in

a hermaphroditic snail, g, the ovo-

testis or hermaphrodite gland, in

which both egg cells and sperma-

tozoa are formed; d, hermaphrodite

duct, through which both egg cells

and spermatozoa pass; ag, accessory

gland; od, oviduct, and sd, sperm

duct, respectively; spt, spermatheca,

in which the sperm of another indi-

vidual is stored after copulation; p,

penis, and m, the muscle retractor of

penis; /, flagellum; Id, "love dart,"

used in the courtship process; mg,

mucous elands.

258 Evolution of Sex

some fishes, in which a young individual first develops and functions

as one sex, for example, as a female, but then changes into a male.

A vestigial hermaphroditism of a similar kind has been found by

Witschi also in some species of frogs.

In hermaphrodites the female and male sex cells, as well as the re-

productive organs in which they are produced, have the same chromo-

somes and genes. At first sight this may seem strange. Why do some

genetically similar cells develop into eggs and others into spermato-

zoa? Why, in dioecious plants, has the same flower ovules and pollen

grains? Actually this is no more remarkable than that some body cells

become, for example, liver cells and others brain cells, although they

have, at least initially, the same genotype. It should be remembered

that in a cell, as in an entire organism, the genotype determines the

reactions to the environment. The fate of a cell may depend upon its

position in the developing body, and on association with other cells.

The male and female parts of a hermaphrodite are simply different

organs of the same body.

Self-fertility and Self-sterility. Since hermaphroditism occurs chiefly

among lower and not at all among higher animals, it is probable that

hermaphroditism is the primitive condition, and that bisexuality arose

later in the process of the evolution of sex. Plato was wrong, after all:

it is the separation of sexes into female and male individuals, not the

hermaphroditism, that may be regarded as the more "perfect" condi-

tion (see above). But the situation is actually too complex for any

such facile judgment. Separation of sexes has no unconditional ad-

vantages over hermaphroditism.

In many hermaphroditic plants, such as wheat, self-fertilization

occurs when the pollen falls on the stigma of the same flower. In

some of these self-pollinating plants the flowers are so constructed

that the pollination takes place before the flowers open and before

the pollen can be transferred to other flowers by wind, by insects, or

by other means. Such an arrangement has advantages as well as

disadvantages. An obvious advantage of self-pollination is that a

regular seed set is assured regardless of the vagaries of weather or of

insects which transport the pollen in most cross-fertilizing plants. To

offset this advantage, continuous self-pollination, like a very close in-

breeding (see Chapter 9), leads to formation of homozygous pure

lines. In the long run, these pure lines deplete the store of hereditary

variability, and rob the species of the chief advantage of sexual re-

production, which is production of new gene combinations. Species

Origin of Bisexuality 259

in which self-fertilization is the only method of reproduction probably

lack evolutionary plasticity.

Many hermaphrodites, therefore, have contrived in the process of

evolution various more or less ingenious adaptations to avoid self-

fertilization. Perhaps the simplest of such adaptations, found in many

plants, is that the male reproductive organs (anthers) mature either

before or after the female organs do. The pollen of a given individual

is thus available only for fertilization of ovules of other individuals.

A more radical method which accomplishes the same function is self-

sterility. Thus in rye, which belongs to a genus (Secale) closely re-

lated to wheat (Triticum), most of the pollen grains fail to germinate

on the stigma of the same plant which produces the pollen. Rye is

an obligatory cross-fertilizer. Self-sterility occurs also in some her-

maphroditic animals. For example, the sea squirt (Ciona) produces

sperm which is incapable of fertilizing the eggs of the same individual,

although there is usually no difficulty in making it fertilize the eggs

of any other individual. Sea squirts in adult condition live attached

to underwater rocks; they discharge their eggs and sperm in the sea

water, and fertilization occurs outside the bodies of the parents. The

sex cells that unite are derived from different individuals.

Origin of Bisexuality. Separation of sexes, bisexuality, is a radical

method to insure that all reproduction takes place by cross-fertilization.

Some individuals become females and produce only eggs, and others

become males and give only spermatozoa. Fertilization involves neces-

sarily a union of gametes coming from different individuals. Bisex-

uality has also the advantage over hermaphroditism that it permits a

division of labor between the sexes. Among animals, males are often

more mobile and active than females; or males furnish protection to

the females and cooperate in raising the offspring; or, simply, the two

sexes carry different organs which it might be difficult to combine in

the body of one individual.

In many animals, though not in all, males actively search for fe-

males, and may be highly polygamous; females tend to be more passive

but more choosy. Beyond doubt, the necessity for the two sexes to

find each other and to perform the actions needed to bring about

fertilization has raised numerous biological problems that could be

solved only with the aid of highly developed nervous systems. Thus

the evolution of mental abilities, culminating in man, might have been

initiated by the separation of sexes.

Organisms are known with various intermediate stages between

hermaphroditism and bisexuality. They make it easy to visualize the

Triploid Intersexes in Drosophila 261

individual or individuals in the pile are females; the topmost indi-

viduals are males. The individuals in the middle are in the process of

changing from the male to the female sex. According to Coe (1936,

1948), the maleness in the limpet is a juvenile characteristic, which

an individual outgrows as it becomes older and as other individuals

become attached to it. The transition from hermaphroditism to bi-

sexuality occurs in the boat shell within the lifetime of an individual.

Chromosomal Sex Determination. In many groups of organisms,

as among insects and vertebrates, hermaphroditism is rare, and the

bisexual condition has become firmly established in the course of evolu-

tion. The reproductive organs and functions of females and males

differ so greatly that any sexual intermediates (intersexes, see below)

are sterile and useless to the species. It becomes, then, important to

have the sex of every individual decided so firmly that it will develop

either as a female or as a male. Any risk of the appearance of sexu-

ally abnormal individuals should be eliminated or minimized. The

sex determination which satisfies the above requirement is by means

of sex chromosomes. The populations consist of two kinds of indi-

viduals, females and males, which differ in their chromosomal com-

plements and have different norms of reaction to the environment.

The pioneer microscopist Leeuwenhoek claimed in 1680 that he

saw two kinds of spermatozoa in the seminal fluid of man, and also in

that of the dog. He conjectured that one kind gave females and the

other, males. The curious thing is that his conjecture was right even

though his observation wrong. More than two centuries after Leeu-

wenhoek, it was discovered that in many animals half of the sperma-

tozoa contain an X-chromosome and the other half a Y-chromosome,

and that the former give rise to females and the latter to males (Chap-

ter 3, pages 55-60). But even under modern microscopes the X-bear-

ing and the Y-bearing spermatozoa cannot be distinguished by their

appearance in the seminal fluids.

The chromosomal theory of sex-determination has elucidated the

problem which excited man's curiosity for many centuries: what makes

seme individuals develop into females and others into males? The

answer is that in most, though not in all, bisexual organisms, sex is

decided at fertilization by the chromosomal constitution which the

fertilized egg comes to possess. Just how the determination of the

normal sexes takes place can be inferred best by studies of some in-

stance of abnormal or experimentally modified sexuality.

Triploid Intersexes in Drosophila. The female flies of Drosophila

melanogaster have two sets of four chromosomes, including a pair of

262

Evolution of Sex

X-chromosomes (that is, a total of 8 chromosomes). The male flies

have one of the X-chromosomes replaced by a Y-chromosome ( Figure

11.5). Around 1920, Bridges discovered some females which were

triploid, that is, had three sets of chromosomes, including three X's

(a total of 12 chromosomes). Triploid females are normal in appear-

ance and behavior, and they produce progeny when mated to normal

diploid males. This progeny consists of ordinary diploid females,

Female

(diploid)

Female

(triploid)

Supetfemale

Intersexes Male Supermale

Figure 11.5. Chromosome complements of the different sexual types in Drosoph-

ila melanogaster. The X-chromosomes are rod-like, and the Y-chromosomes hook-

shaped.

diploid males, triploid females, and three new sexual types known as

intersexes, superfemales, and supermales.

The origin of these sexual types becomes clear if the meiosis in their

triploid mother is considered. Let us distinguish between the X- and

Y-chromosomes on one hand and the autosomes on the other—the

latter being chromosomes which are alike in females and in males. A

triploid female carries, then, three X-chromosomes and three sets of

autosomes. It produces four kinds of eggs: (1) carrying one X-

chromosome and one set of autosomes, (2) carrying one X and two

sets of autosomes, (3) carrying two X-chromosomes and two sets of

autosomes, and (4) carrying two X's and one set of autosomes (there

are also some inviable eggs produced which need not be considered

here). A normal male produces, of course, two kinds of spermatozoa,

with an X- or a Y-chromosome, and with a single set of autosomes.

Triploid lntersexes in Drosophila 263

The fertilization yields, therefore, the eight kinds of offspring, which

are shown in Table 11.1.

TABLE 11.1

SEXUAL TYPES IN THE PROGENY OF TRIPLOID DROSOPHILA

X-Chromosomes Autosomes

Sets of

Y-Chromosome Ratio

Sex

(X)

(A)

(Y)

X:A

Superfemale

9

9

* * «

1.50

Triploid female

S

3

• • •

1.00

Diploid female

%

2

• ' '

1.00

Diploid (XXY) female

*

t

1

1.00

Intersex

2

3

' * •

0.67

Intersex

2

9

1

0.67

Male

1

2

l

0.50

Supermale

1

9

1

0.33

Before the discovery of triploid females and their progeny, it was

possible to entertain the hypothesis that the X-chromosome contains

genes which impel the development towards femaleness, and that the

Y-chromosome is the seat of genes which tend to maleness. Thus,

XX = female, and XY = male. But this hypothesis is contradicted by

the fact that individuals which have a Y-chromosome in addition to

two X-chromosomes are normal females, provided that they have two

sets of autosomes (Figure 3.14). Bridges concluded that, in Dro-

sophila, the maleness is carried not in the Y-chromosome but in the

autosomes, and the femaleness in the X-chromosomes. Sex is deter-

mined by a "balance" of the tendencies towards femaleness and male-

ness. This "balance" is decided by the ratio of the number of X-

chromosomes and of sets of autosomes which an individual carries.

This ratio for the various sexual types is shown in the rightmost col-

umn of Table 11.1.

The ratio is unity both in diploid and in triploid females, although

the former have two and the latter three X-chromosomes. The ratio

Diploid Intersexes 265

extreme than the normal sexes. The Y-chromosome carries, in Dro-

sophila, no sex genes. Its presence or absence fails to modify the

sexual characters as set by the other chromosomes.

Polyploids in Melandrium. The fact that in Drosophila the female-

ness is carried in the X-chromosome and the maleness in the autosomes

does not mean that this is so in all bisexual organisms. The same

biological problem, sex determination, has been solved in different

ways in different organisms. As an example of such diversity of ways

we may consider the studies—carried out by Warmke and Blakeslee

in the United States and by Westergaard in Denmark—of sex de-

termination in the plant Meltindrium album.

In this plant some individuals produce only female flowers and

others only male flowers. Plants of both sexes have 22 autosomes (11

pairs), females have in addition a pair of X-chromosomes, and males

one X- and one Y-chromosome. The investigators obtained also triploid

(33 autosomes) and tetraploid (44 autosomes) plants, with one to four

X-chromosomes and with none, one, or two Y-chromosomes. In con-

trast to Drosophila, Melandrium proved to be female regardless of the

ratio between the numbers of X-chromosomes and sets of autosomes,

provided only that the Y-chromosome is absent.

Individuals with two X-chromosomes and three sets of autosomes

are pure females in Melandrium and intersexes in Drosophila. Again

contrasting with Drosophila, plants which carry one or more Y-chromo-

somes are either males or hermaphrodites (which have flowers with

both pollen and ovules). Diploid, triploid, and tetraploid plants are

male when they carry one X and at least one Y. When two or three

X-chromosomes are present, one Y-chromosome gives male plants with

an occasional hermaphroditic flower, but two Y-chromosomes cause

the plant to be a pure male. A tetraploid plant with four X's and one

Y is a hermaphrodite.

It is clear from this evidence that in Melandrium the Y-chromo-

some carries a gene, or genes, for maleness. The X-chromosome is

the carrier of femaleness; the autosomes seem to be sexually neutral.

It should also be noted that in Drosophila sexual unbalance gives

sterile intersexes. In Melandrium the sex intergrades are plants with

more or less perfect hermaphroditic flowers, which are at least par-

tially fertile.

Diploid Intersexes. The method of determination of sex which has

proved most successful in evolution is by means of sex chromosomes.

In Drosophila the sex of an individual is decided by the number of

X-chromosomes in relation to the number of autosomes; in Melandrium

Diploid lntersexes 267

(Lymantria dispar) described by Goldschmidt in numerous publi-

cations from 1911 to 1938. In this insect, the females and males differ

strongly in coloration (Figure 11.7) and in behavior. The fen\ale is

sluggish; the male is an agile flier. The species occurs in most of

Europe and in northern Asia—from the Atlantic Ocean to Japan. It

was also introduced, and became a pest, in the United States. When-

ever female and male gypsy moths from the same locality in this

enormous territory are crossed, the sex-determining mechanism func-

tions perfectly—the offspring consists of normal females and males.

But crosses between moths from Japan and those from Europe produce

some normal females or males and also some intersexes.

The details of this case are complex, but the main features of the

story are clear enough. All races of the gypsy moth are diploid, and

in all of them the male has two X-chromosomes and the female one

X- and one Y-chromosome. The X carries the gene (or genes) for

maleness, whereas the femaleness is carried, according to Goldschmidt,

not in the chromosomes at all but in the cytoplasm. The situation,

then, is quite different from that observed in Drosophila or in Melan-

drium. Within the population of any one geographic region, the

genetic determiners of femaleness and of maleness have become mu-

tually adjusted by countless generations of natural selection. The

result is that two doses of maleness brought in by two X-chromosomes

overpower the femaleness of the cytoplasm, and produce a normal

male. Yet one dose of maleness yields to the femaleness, and an XY

individual is always a normal female.

The "strength" of the sex determiners in the race living in Japan is

greater than that of the sex determiners in European races. An X-

chromosome of the Japanese race meets, in the hybrids between the

races, the cytoplasm and the Y-chromosome of European origin. The

maleness of the zygote is, then, too strong to submit to the female

tendency, and the result is an XY individual which is not a female

but an intersex. On the other hand, a race hybrid may also have two

X-chromosomes with "weak" European maleness, opposed by a "strong"

Japanese femaleness in the cytoplasm. The result is an XX individual,

which is not a male but an intersex.

Diploid intersexes teach us two important lessons. First, sex is not

a rigidly fixed quality but a variable quantity. There may be more

or less femaleness and maleness in an organism. Second, the strength

of the sex determiners is a matter which is so adjusted by natural

selection that in all environments to which a species is normally ex-

posed the sex mechanism functions smoothly. Either females or males

268 Evolution of Sex

are produced, and sex intergrades which are useless to the species are

avoided.

Se* and Environment. Many of the spurious solutions of the prob-

lem of sex determination which were in vogue before the discovery

of sex chromosomes assumed that the sex of an individual is decided

by the environment in which this individual develops. Such notions

still crop up from time to time, especially in the popular press.

Can the environment influence sex? We know that most bisexual

organisms produce either two kinds of spermatozoa (mammals, most

insects), or two kinds of eggs (birds, butterflies, and moths), and that

the chromosomal sex of an individual is decided at fertilization. It

is, then, conceivable that methods will be found (though none have

been found so far in higher animals) which will favor fertilization of

eggs, for example, by male-determining spermatozoa and will dis-

criminate against female-determining ones, or vice versa. On the other

hand, what is fixed in the zygote at fertilization is its genotype. But

the genotype sets only the norm of reaction of the organism to the

environment, and does not necessarily decide any characters or struc-

tures. One may distinguish the sexual genotype from its actual mani-

festation, the sexual phenotype. Sexual traits of an individual may,

then, depend on the environment as well as on the genotype. Modi-

fication of the sexual phenotype by the environment is, in principle,

quite possible.

Triploid intersexes in Drosophila (see above) have two X-chromo-

somes and three sets of autosomes. Yet, despite their chromosomal

uniformity, these intersexes are quite variable in sexual characters

(Figure 11.6). Some of them resemble normal males in behavior and

in the structure of reproductive organs; others are female-like. Still

others have mixtures of malformed organs of both sexes. In short,

all transitions between female and male states can be found among

the intersexes.

What causes these great variations in the sexual phenotype? The

environment has a great deal of influence in this matter. When triploid

females are crossed to normal diploid males, the progeny consists of

several sexual types, including intersexes (see Table 11.1). Cultures

of triploid females were exposed to four different temperatures—15°,

20°, 24°, and 28°C. In cultures which developed at the lowest tem-

perature most of the intersexes developed male-like characters. Con-

versely, at high temperatures the intersexes proved more female-like.

Cold turns the development of Drosophila towards masculinity, and

heat towards femininity. Changes in the sexual phenotype of the

Sex and Caste in Hymenopteran Insects 269

intersexes can also be produced by selection. If we select the triploid

females which yield, in a given environment, most female-like, or most

male-like, intersexes, strains are soon established which differ in their

sexual tendencies.

Factor of Safety. The experiments concerning the influence of the

environment on intersexes are made by exposing the progenies of

triploid mothers to different temperatures or other influences. Now

their progenies contain not only intersexes but also diploid females

and males, triploid females, superfemales, and supermales. An aston-

ishing fact, at first sight, is that the temperature variations which alter

greatly the sexual phenotype of the intersexes have apparently no

effects on the females and males. The sexual development of females

and males is homeostatic (see page 13), that is, it is so buffered against

environmental disturbances that a female chromosomal constitution

always yields a normal female phenotype, and male chromosomes

yield a normal male. The developmental pattern of the intersexes,

on the contrary, is set so precariously that it is easily swayed towards

either femaleness or maleness by both environmental and by genetic

changes.

The reason for the more perfect homeostatic adjustments in females

and males than in intersexes is not far to seek. The whole existence

of the species depends upon the reproductive performance of the

females and males of which it is composed. Natural selection, accord-

ingly, has built in a "factor of safety" in the norms of reaction of

females and males. This "factor of safety" operates in such a manner

that in all environments to which the species is exposed the develop-

ment always culminates in either femaleness or in maleness. The

absence of the "factor of safety" in the intersexes is also understand-

able. Intersexes are rare or absent in nature, and their norm of

reaction has not been perfected by natural selection. It is poised

between femaleness and maleness, and can be turned towards or away

from either.

Sex and Caste in Hymenopteran Insects. Once we understand that

sex is a form of evolutionary adaptation, the many curious variations

of the sex-determining mechanisms found in various creatures become

intelligible. The same end may be attained by different means, and

any method of sexual reproduction which safeguards the procreative

capacities of the species may become perfected and established in

evolution.

Bees, wasps, ants, saw flies, parasitic and gall wasps constitute the

insect order Hymenoptera, which is generally regarded as the most

270 Evolution of Sex

progressive among insects. However, sex chromosomes of the usual

kind are missing in this order. Instead, eggs which remain unfertilized

develop parthenogenetically into males. Eggs of the same mother

which' are fertilized by spermatozoa develop into females. Thus

females have twice as many chromosomes as males; the former are

diploid and the latter haploid. Virgin females deposit eggs which

yield only sons; females impregnated by males deposit some eggs

which yield daughters and others which give sons. It appears that a

female can control the access of spermatozoa to some of her eggs and

can withhold them from others.

Where sex is determined by sex chromosomes (the XX-XY mech-

anism) the number of female and male zygotes that are formed is,

with some rare and little understood exceptions, not far from equality.

The diploid-haploid sex mechanism of the Hymenoptera permits the

progeny to contain any proportions of sons and daughters. Females

often greatly outnumber the males, especially among social insects,

such as ants.

An ant colony contains several categories of individuals. An ant

nest usually has numerous "workers," and in some species also "sol-

diers," which are diploid individuals, and hence genetically females.

Their reproductive organs, however, remain underdeveloped. A rela-

tively small number of sexually developed females and males suffice

to take care of the perpetuation of the species. The males have

haploid chromosome complements, whereas the sexually developed

females ("queens") and the workers are chromosomally identical, hav-

ing diploid chromosome complements. Whether a diploid zygote

develops into a queen or into a worker depends, at least in the domes-

tic honey bee and in some ant species, entirely upon the kind of food

which the workers give to the growing larvae in the honeycomb or

in the ant nest. The diploid chromosome complement thus sets a

norm of reaction which permits the development of either a worker

or a queen. What develops from a given egg is decided by the en-

vironment. However, the Brazilian geneticist Kerr has found that in

stingless bees of the genus Melipona, the difference between queens

and workers is brought about by their genotypes. Different repro-

ductive mechanisms have arisen in evolution even in such closely

related forms as different genera of bees.

Sex in Ophryotrocha and in Bonellia. We have seen above that in

the boat shells the sex of an individual varies with its age and its

position in the colony. Hartmann and his students (1936, 1938) found

a somewhat similar situation in the marine worm Ophryotrocha puer-

ilis. All young individuals of this worm are males, but as the worms

274 Evolution of Sex

the classical work of Bridges and Morgan (1919). Gynandromorphs

begin their existence as female zygotes, having, of course, two X-

chromosomes. Normally both X's are reduplicated at cell divisions,

and are inherited by all cells of the developing embryo. But about

once in some tens of thousands of eggs (and more often under the

influence of X-ray treatments) one of the X-chromosomes may be acci-

dentally lost, and a cell is formed which contains a single X-chromo-

some (and two normal sets of autosomes). A single X-chromosome

results in a male chromosome complement. All the cells which arise

by division from the cell that has a single X-chromosome are, then,

male cells. A part of the body of a gynandromorph, therefore, is male,

and the remainder is female. The relative sizes of the female and male

parts depend, of course, on whether the X-chromosome is eliminated

early or late during the development of the egg.

The important fact is that, in Drosophila, the dividing line between

the female and male fractions of the body of the gynandromorph is

usually sharp. Even when the male fraction is small and is completely

surrounded by female tissues, or vice versa, the sexual characteristics

are of one sex or the other. The sex of a tissue is independent of

contacts with cells of the same or of the opposite sex.

Sex Hormones in Vertebrates. About a century ago the great physi-

ologist Claude Bernard (1813-1878) pointed out that in some glands

the secretions leave the gland by way of a duct, whereas other glands

secrete their products in the blood stream. The latter are called

glands of internal secretion or endocrines. Their secretions are hor-

mones, chemical messengers which are carried to various parts of the

body, where they influence the activities of cells and tissues.

Adrenalin, a hormone of the adrenal gland, was discovered by Abel

in 1898, and secretin, prepared by certain intestinal glands, was found

by Bayliss and Starling in 1902. Since then, numerous hormones have

been discovered and studied in more or less detail; more than twenty

compounds are produced by the cortex of the adrenal gland alone.

Many hormones have been isolated, analyzed, and synthesized arti-

ficially. Thus adrenalin (or epinephrin) was isolated in 1901 and syn-

thesized in 1904; thyroxin isolated in 1916 and synthesized in 1927;

testosterone crystallized and synthesized in 1935, etc. Hormones may

operate in incredibly low concentrations; epinephrin, for instance, is

active in a dilution of 1 : 300,000,000.

Hormones exist not only in the higher but also in the lower animals

and in plants. But it is in the vertebrates that they have become most

Sex Hormones in Vertebrates 275

important as coordinators of physiological functions, and particularly

in the manifestations of sex. Furthermore, the effects of many endo-

crine glands in mammals and in man are so closely interlocked that

the system of endocrines functions as an interdependent whole. Thus

a deficiency or an excess of the hormones of the pituitary gland affects

the functions of the thyroid gland, of the adrenal cortex, of production

of insulin, of male and female reproductive glands, of parathyroids,

of the rate of growth of the body, etc.

It has been known since antiquity that castration of boys results in

lack of development of many secondary sexual characters, for ex-

ample, of masculine voice, of facial hair, masculine body shape, and

male sexual drive. Similar changes have long been known in castrated

males of other animals, such as capons, geldings, and oxen (castrated

roosters, stallions, and bulls respectively). These changes are caused

primarily by the absence of testosterone, a hormone produced by the

interstitial cells of the testicles (not by the cells which give rise to

spermatozoa!). Testosterone belongs to the class of compounds known

as sterols, and is related to cholesterol, which is present in many ani-

mal tissues.

In the adult human female an egg is shed, usually from either the

right or the left ovary, once in about 4 weeks. During the repro-

ductive age, from the first menstruation to menopause, a total of some

400 eggs is produced. Each egg cell grows in a so-called Graafian

follicle, which is a liquid-filled vesicle under the surface of the ovary.

The Graafian follicles are apparently the source also of the hormone

estrogen. Estrogen stimulates the development of the lining and

glands of the inner surface of the uterus. After the egg is shed, the

Graafian follicle develops into a glandular structure of a different kind,

the corpus luteum. This structure secretes a hormone progesterone,

which causes further changes in the lining of the uterus, and assists in

the implantation of the fertilized egg if pregnancy is achieved.

The estrogen hormone was isolated by Doisy in 1929 and synthesized

by Butenandt in 1930. Progesterone was isolated and synthesized by

Butenandt in 1934. The chemical structures of both hormones are

related to cholesterol and to the masculine hormone, testosterone

(Figure 11.11). However, their effects are, of course, quite different.

The estrogens play an important role in the development of female

secondary sexual characters, such as feminine body shape, voice, and

feminine sexual drive.

Pezard (1918), Zavadovsky (1922), and many investigators since

then have experimented with castration and with transplantation of

276

Evolution of Sex

gonads (testes or ovaries) in birds, particularly in poultry. Castrated

roosters (capons) retain most of their "masculine" plumage, but the

combs and wattles shrink and lose their turgidity, the animal ceases to

crow, and loses its pugnacity and its sexual drive. Castrated hens

acquire, after a molt, a capon-like plumage, and come to resemble in

appearance the castrated males. The plumage of a capon and under-

development of combs and wattles are thus sexually neutral characters;

male plumage is suppressed by the hormones secreted by the ovaries,

whereas testicular hormones stimulate the development of combs,

wattles, and spurs. Implantation into castrated males and females

Te*toet«roDe

Eatrogene

(oc»trone)

Progntenme

Figure 11.11. Chemical structure of three sex hormones.

of gonads from other individuals produces the characteristic feminiz-

ing or masculinizing effects, and these regardless of what the chromo-

somal sex of the individual may be. An implanted ovary causes the

development of female plumage; implanted testes give cock's comb

and wattles, together with behavior characteristic of the sex of the

implanted gonad. However, the larger body size of roosters com-

pared to the hens seems to be independent of the hormones.

Hormonal Intersexuality. Castration, transplantation of gonads,

and injection of sex hormones obviously do not change the chromo-

somal sex of the individual. It is fallacious to speak of the effects on

sex of genes and of hormones as though they were alternatives, which

they are certainly not. The hormone system is a part of the develop-

mental mechanism whereby the genotype becomes realized in a cer-

tain sexual phenotype. The functioning of the hormone-producing

glands is itself determined by the genes; it is a part of the phenotype.

Castration and hormone injections are environmental agencies.

Hormonal disturbances result, in man and in other animals, from

many pathological conditions, caused in turn either by defective

heredity or by environmental mishaps. Cases of sexual intermediacy

Sexual Selection 277

were known even in antiquity, and some of them were immortalized

in ancient art. Instances of individuals who "changed their sex" are

reported from time to time in medical literature and in the daily press.

More or less masculine women and woman-like men are not un-

common.

Some of the alleged "changes" of sex are simply cases where the

sex was incorrectly determined at birth owing to malformation in the

external sex organs. Others are instances of hormonal intersexuality.

Hormonal intersexuality may be due either to developmental accidents

or to genetic causes. At least one group of human pedigrees from

Sweden, recently described by Pettersen and Bonnier, suggests a domi-

nant mutant gene which has no effects in females (individuals with

two X-ehromosomes) but which transforms the males (XY individuals)

into intersexes. The situation is comparable to diploid intersexuality

in Drosophila (see page 266; in Drosophila it is, however, the genetic

females which become intersexual).

Lillie, Tandler, and Keller (1916) have furnished a very clear

analysis of one kind of developmental accident. It had been known

for a long time that when a cow gave birth to a pair of twins of unlike

sex, the female calf, called a freemartin, was often sexually abnormal

and sterile. Freemartins have underdeveloped external genitals of

female type, but the internal reproductive organs are intersexual,

verging towards maleness. Lillie and his colleagues found that this

condition occurs when the fetuses share a common placenta and

common blood circulation which are often established between the

twin embryos during pregnancy (Figure 11.12). Freemartin is the

female member of a pair of fraternal twins (which arise from two sep-

arate eggs fertilized by different spermatozoa). The chromosomal

constitution of the freemartin is certainly XX, but the sexual develop-

ment is interfered with by the hormones received through blood from

the male co-twin. A connection between the blood circulations of the

two embryos may be established quite early during the pregnancy,

and the male hormones may upset the development of the female twin

in its most sensitive stages.

Sexual Selection. In 1871 Darwin published his theory of sexual

selection, to account for the origin in evolution of conspicuous sec-

ondary sexual characters, such as the antlers of male deer, the bright

plumages and elaborate displays and songs of many birds, the lion's

mane, and facial hair growth in the human male. Since these charac-

ters, according to Darwin, are not sufficiently useful to the species to

be formed by natural selection, he believed that their usefulness lay

Sexual Behavior 279

mechanisms between animal species. Related species may look differ-

ent, have different voices, and act differently in part because these

differences are recognition marks that enable individuals to identify

each other as members of the same species. These recognition marks

may reach almost incredible degrees of complexity.

In many species of moths males locate females by scent. Experi-

ments of Fabre (1823-1915), one of the keenest students of insect

behavior, showed that males may be attracted from considerable dis-

tances, and may overcome various obstacles to reach a female con-

fined in a cage. The scent, quite imperceptible to the human nose, is

strictly specific, so that males of wrong species are seldom or never

attracted. In other animals, including some insects, vision is more

important than smell in finding a mate. This is probably true in many

birds. In still others, sound stimuli are important. The "songs" of a

bird, a cricket or a locust, the croaking of a toad, and the buzzing of

at least some mosquitoes are among the methods whereby the sexes

advertise their presence and find each other. How different may be

the requirements of even very closely related species is shown by the

flies Drosophila pseudoobscura and Drosophila subobscura. The lat-

ter species copulates only in the presence of light; the former copulates

in the light or in the dark apparently indiscriminately.

Sexual Behavior. In many animals the behavior patterns connected

with reproduction became complex "rituals" of courtship and mating.

In a species of the fly family Empidae a courting male alights near a

female and vibrates his wings. The vibration probably gives an

auditory stimulus, and eventually the female responds by a similar

vibration. The male now lifts up his front legs and waves them

about, moving closer and closer to the female until they "caress" each

other with their uplifted front legs; a tactile or a chemical stimulation

seems to be involved in this "caressing." In a different species of the

same family, the male catches another insect as a prey (these flies are

predators); he then swaddles the prey in silk threads, seeks out a

female, and presents the prey to her. She accepts the "gift," and

mating takes place while she is busily unwrapping the food. The

presentation of the "gift" is in some species, but not in others, accom-

panied by a special "dancing ceremony"; the dancing may be done by

the male alone, or both sexes may take part.

Among birds, the greatest complexity of sexual behavior is reached

probably in Australian bower birds. Here the male builds a kind of

shelter consisting of boughs and twigs. Then he proceeds to "dec-

orate" the shelter with colored materials, such as flower petals, feath-

Sex Reversal 281

precise causation is little known. Increasing complexities of human

societies may at times lead to misdirection of the sexual urge. On the

other hand, there is some evidence that certain forms of homosexu-

ality may have a genetic basis. Kallmann and his collaborators (1953)

found that, among 44 pairs of male identical twins one member of

which had shown overt homosexual behavior as an adult, the co-twin

in every case also evinced behavior of this kind. And yet, among

51 pairs of fraternal (two-egg) twins where one of the twins was

known to be a homosexual, the co-twin showed evidence of similar

condition in only 13 cases. It appears, then, that some human geno-

types react by evolving homosexual behavior in at least some social

and cultural environments.

Sex Reversal. A crucial proof that the genotype merely sets the

stage for the development of a certain sex in a certain range of environ-

ments is afforded by sex reversals. Males of some species of toads

have, next to the testis, a so-called organ of Bidder (Figure 11.13),

which under the microscope reveals a structure resembling an ovary.

The function of the organ of Bidder in normal males is unknown.

But Harms, Ponse, Witschi, and others have shown that when a male

toad is castrated (that is, when the testes are removed by a surgical

operation), the organ of Bidder increases in size and becomes a

diminutive ovary in which a certain number of fertilizable eggs are

produced. The castrated male now functions as a female, and may

produce progeny when mated to a normal male.

A castrated male must conserve, despite its sex reversal, its normal

chromosomal composition. This can be proven by observing the sex

distribution in the offspring of the cross, castrated male X normal

male. Miss Ponse found this offspring to consist entirely of males,

and interpreted the result to mean that in toads the male sex has two

X-chromosomes (XX) and the female sex an X- and a Y-chromosome.

All the sperms of a normal male, and all the eggs of a male transformed

into a female will then carry an X-chromosome, and the experimental

progeny will be purely masculine. The experiments of Harms, how-

ever, gave an entirely different result: about one-quarter of the eggs

died, half of the eggs developed into males, and one-quarter into

females. He interprets this to mean that a male toad had an X- and

a Y-chromosome, and consequently that two kinds of eggs and two

kinds of sperms were produced by his experimental animals. If so,

half of the resulting zygotes will be XY (males), one-quarter XX

(females), and one-quarter YY (these are inviable). The results of

Ponse and Harms are not necessarily contradictory, since different

Suggestions for Further Reading 283

Koch, F. C., and Smith, P. E. (Editors). 1942. Sex hormones. Biological

Symposia, Volume 9. Jaques Cattell Press, Lancaster.

Robson, J. M. 1947. Recent Advances in Sex and Reproductive Physiology. 3rd

Edition. Blakiston, Philadelphia.

The literature on sex hormones is enormous, but it is reviewed very well in

the three books just cited.

Burton, M. 1954. Animal Courtship. F. Praeger, New York.

A well-written review of courtship habits in various animals, but the interpreta-

tion given is out of tune with modern evolutionism.

Wenrich, D. H. (Editor). 1954. Sex in Microorganisms. American Association

for the Advancement of Science, Washington.

This is a symposium on sexual processes in lower organisms.

Those familiar with the German and French languages will find also the fol-

lowing two books most useful:

Hartmann, M. 1947. Allgemeine Biologic. 3rd Edition. Gustav Fischer, Jena

Canllery, M. 1951. Organisme et sexualite. 2nd Edition. G. Doin, Paris.

12

Historical Record of

Organic Evolution

Almost one and one-half centuries ago (1809) Lamarck compared

the process of evolution to the movement of the hour hand of a clock.

A creature whose lifetime lasted one second would perceive no mo-

tion at all in the hour hand. Even thirty consecutive generations of

such a creature might doubt whether the hand really moves. With

respect to observing evolution in geological time, man is in the posi-

tion of such a creature.

Since Lamarck, biologists did find ways to perceive some motion in

the hour hand of the evolutionary clock. They used for this purpose

the speeded-up processes of evolution by polyploidy, evolution under

domestication (Chapter 9), and the experimental evolution, chiefly in

microorganisms (Chapter 5). But with respect to the evolution which

has actually taken place in the history of the earth, an observer of

only the now-living animals and plants is still in a position of judging

a long movie film by only the last picture frame. Let us, though, give

the credit which is due to this observer: from the only picture at his

disposal he has correctly inferred that there was a story back of this

picture. Moreover, he has successfully used the experimental method

to find out some of the mainsprings of action in the story. He can

even make intelligent guesses as to what kinds of events could and

could not have occurred. But to learn what events did actually occur

and in what sequence, the last picture frame is not enough. Fortu-

nately, some fossil remains of the living beings of the past epochs are

preserved in the geological strata. They furnish us with some torn-up

fragments of the preceding picture frames of the story. It is the busi

ness of paleontologists to restore these fragments as much as possible,

and to arrange them in the proper sequence.

284

The Fossil Record Is Incomplete 285

The Fossil Record Is Incomplete. Evolution is a continuous proc-

ess, composed, though, of small discontinuous mutation steps. From

the continuity of evolution it follows, of course, that forms interme-

diate between the now-living organisms must have existed in the past.

Provided that man and the simplest virus are really descended from

common ancestors, then, if all organisms which lived in the past were

fossilized and recovered, we would see an unbroken chain of organ-

isms intermediate between man and the virus. (This assumes, of

course, that the entire living world is monophyletic, descended from

a single common ancestor.)

In reality, only an infinitesimal fraction of individuals who lived in

the past have become preserved as fossils. The fossil record is incom-

plete by the very nature of things. The remains of a creature are pre-

served in the sediments that form the stratified rocks of the earth's

crust only as a result of rare accidents. Soft-bodied organisms are

preserved only quite exceptionally, although some of them did leave

their shadow pictures as prints in the rocks. For organisms which

possess hard parts, such as solid calcareous shells or skeletons or

teeth, the chances of preservation are relatively much greater, though

still very small in the absolute sense. Moreover, fossilization of the

remains of a creature is not enough. Many sedimentary rocks, which

doubtless contained fossils, were eroded away at various stages of the

earth's history, and the fossils were thus irretrievably lost. Very nat-

urally, more ancient fossils were lost in this manner more often than

relatively more recent ones. Other old rocks were buried under enor-

mous weights of newer sediments, and consequently subjected to very

high pressures and high heat. This heat has led to the rocks being

metamorphosed and recrystallized, and again to the loss of the fossils

which the rocks had contained. Ancient rocks are metamorphosed, of

course, more frequently than recent ones. The paleontologists have

thus far discovered and studied only an infinitesimal fraction of the

fossils that are preserved.

In view of all this, it is not surprising that the known fossil record

is nowhere near sufficient to reconstruct even approximately the phylo-

genetic relationships of all organisms. In fact, so obvious is the in-

completeness of this record that Darwin expended rather more effort

in his Origin of Species to show that paleontology does not contradict

evolution than he did to show that it proves evolution. Since Darwin,

the unrelenting efforts of paleontologists have filled many gaps in the

fossil record, but many more remain to be filled. In some particu-

larly favorable instances the fossils are so abundant that the changes

Geological Time

287

TABLE 12.1

THE GEOLOGICAL PERIODS AND THE GEOLOGICAL TIME SCALE

(After Colbert.)

Duration

in Millions

of Years

Millions of

Years Since

Beginning

Eras

Periods

Epochs

Quaternary

1 Recent

I Pleistocene

0.025

1

0.025

1

Cenoaoic

Pliocene

10

11

Miocene

15

26

Tertiary

Oligocene

Eocene

10

19

36

55

Paleocene

15

70

Cretaceous

60

130

Mesoaoic

Jurassic

35

165

Triassic

35

200

Permian

30

230

Paleozoic

Pennsylvanian

Mississippian

Devonian

20

30

50

250

280

330

Silurian

30

360

Ordovician

70

430

Cambrian

90

520

288 Historical Record of Organic Evolution

Most Ancient Life. Until the Devonian time, that is, from Cam-

brian through Silurian periods (Table 12.1), all the fossils known are

remains of water-dwelling organisms, and chiefly of sea life. This is

in agreement with the inference that life has originated in the sea,

reached by evolutionists even before Darwin. The basis of this infer-

ence was that most of the now-living primitive organisms are water

dwellers and particularly sea dwellers. Recent physiological studies

have brought an unexpected corroboration of this view. The salt con-

centration in the blood and other body fluids of land animals is ap-

proximately like that in sea water and in marine animals. It is as

though land animals still lived in the sea as far as their internal en-

vironment is concerned. It seems, then, quite natural that remains of

land life appear in the fossil record some 150 to 200 million years later

than do remains of marine life.

Another fact which seems appropriate from the evolutionary point

of view is that the most highly organized phylum of animals, the verte-

brates, are not represented at all in the Cambrian rocks. The verte-

brates appear in the Ordovician period (see Table 12.1), and the first

remains of the vertebrates are those of jawless fishes. This is the most

primitive class of the vertebrates now living. (The best-known mod-

ern representative is the lamprey.)

But beyond the above two features, Cambrian remains do not strike

us either as particularly primitive or lacking in diversity. These re-

mains include representatives of most of the major groups of animals

which live in modern seas. This means that at least some of the Cam-

brian creatures were quite elaborate and advanced in body structure.

Such a situation may seem to be just about the reverse of what was

expected at least by the nineteenth century phylogenetic school of

evolutionists (see Chapter 10). They liked to picture the hypothetical

common ancestors of the now-living organisms as lacking the special

characteristics which these latter have, in other words as "types" or

"ground plans" of the modern creatures. But evidently the seas and

the continents of the past were not inhabited by mere schemes; these

inhabitants were adapted to cope with their environments, just as

modern life is adapted to modern environments.

Quite evidently, the Cambrian fossils represent organisms which

were nowhere near the first to inhabit the earth. The origin of life

from inanimate matter must have taken place very much earlier, and

quite a lot of evolution must have intervened before the Cambrian sea

dweller could have arisen. There is no doubt but that the earth was

more than four billion, and possibly more than five billion, years old

The Age of Amphibians and the Age of Reptiles 289

when Cambrian rocks were formed, but this enormous time span has

left almost no fossil record (see Chapter 1, page 2). The pre-Cam-

brian rocks are mostly metamorphosed, and as such devoid of fossils.

The exceptional non-metamorphosed rocks have yielded disappoint-

ingly little. It is possible that the remains of the most ancient life are

lost forever; at any rate they have not been discovered.

Extinction and Replacement. Not even a brief group-by-group

description of the fossil record of all forms of life can be given in this

book. We must limit ourselves to consideration of only some exam-

ples which illustrate certain general principles.

One of the most striking facts of paleontology is that the inhabitants

of a given period are descended not from all the inhabitants of prior

periods, but from only a part of them. In other words, many creatures

leave no descendants at all; their race becomes extinct. As a matter

of fact, the most probable fate of any group of animals or plants in

the course of time is extinction. If we examine the fossils of any re-

mote geological age, say Mesozoic or Paleozoic, we find that most of

them represent organisms which have no direct descendants living at

present. They have succumbed at some point in time. On the other

hand, some few of the denizens of the past have multiplied greatly,

and their descendants, usually modified by evolutionary changes, fill

the earth today. For one of the things which the fossil record shows

beyond reasonable doubt is that the diversity of life has become

greater and greater in the course of time. The dying-out groups of

organisms are replaced not only by the surviving ones, but also by

quite new organisms, which exploit the environment in novel ways,

and are added from time to time. The known history of the land-

dwelling vertebrate animals illustrates well the phenomena of extinc-

tion and replacement.

The Age of Amphibians and the Age of Reptiles. The most primi-

tive backboned animals that live on land belong to the class Amphibia,

of which frogs, toads, and salamanders are the best-known living rep-

resentatives. The most ancient known amphibians have left some

fossils in the late Devonian deposits of Greenland. The ancestors of

these early amphibians apparently belonged to a group of fishes

(crossopterygians), some of which inhabited fresh-water pools and

streams of this remote age. It is a plausible guess that the develop-

ment of the lungs for air breathing, and the development of walking

legs in the place of the fins, occurred in response to a periodic drying-

up of these pools and streams. Such periodic streams can still be

observed in countries where rainy and dry seasons alternate. It is

290 Historical Record of Organic Evolution

evidently a great advantage for the stream inhabitants in such cli-

mates to be able to respire both in water and in the air, and to be

able to walk overland in search of bodies of water that have not dried

up. Once provided with lungs and with walking legs, the ancient

amphibians became able to exploit also a new source of food, which

was not available to their water-dwelling ancestors, namely, the in-

sects which were becoming numerous at about the same time. We

see that the lungs and legs were "evolutionary inventions" which were

made probably in response to peculiar aquatic environments—period-

ically drying-up pools and streams. But these same "inventions,"

so to speak, opened inadvertently much wider possibilities of adap-

tation to quite novel land environments hitherto unexploited by ver-

tebrate animals.

However that may be, the amphibians have become very numerous

and diversified in the Pennsylvanian period, some 30 to 40 million

years after their first known appearance in the Devonian. Many strik-

ingly large and powerful forms appeared, some of them resembling

modern crocodiles in size and possibly in habits. Such rapid rises in

abundance and diversity of novel groups of organisms adapted to

novel ways of life are sometimes described as "evolutionary explo-

sions"—a description more dramatic than accurate in view of the im-

mense time spans actually involved.

The dominance of amphibians among the land life continued into

the Permian period, but it was soon surrendered to another set of

newcomers—the reptiles. Most of the ancient types of amphibians

died out without issue. Some few of them, however, clung to life, and

eventually produced new groups—frogs and toads, which are reason-

ably successful today but were rare or did not yet exist during the

Age of Amphibians.

The reptiles may be described as the most successful descendants

of a small group of ancient amphibians. It is reasonably clear which

evolutionary invention caused their success. Amphibians are unable

to live entirely on land, since they pass through larval stages (tad-

poles, pollywogs) which live in water. It is interesting that a total

emancipation from the necessity of beginning their lives in water has

been attempted in some toads, in which the tadpoles develop in special

skin pouches on the back of the parent. But, of course, it is the ap-

pearance of eggs provided with abundant food for the development

of the embryo, and with a shell to prevent drying, that solved the

problem of life on dry land. The reptiles, and among their descend-

ants the birds and some few mammals, lay such eggs.

Adaptive Radiation 291

The first reptiles have been found in the Pennsylvanian deposits,

and during the Permian period the reptiles became numerous. They

increased in diversity and in importance during the Mesozoic Age of

Reptiles, reaching a climax in the Cretaceous period. A great crisis

came in late Cretaceous and early Tertiary times, when most groups

of reptiles died out and were replaced by mammals and birds.

Adaptive Radiation. The skeletons of the earliest known reptiles

(cotylosaurs) were so amphibian-like that, in the absence of informa-

tion on what kind of eggs these animals laid, assigning them to either

class is rather arbitrary. Such clear transitions between different

classes and other large groups of animals and plants are so rare in the

fossil record that their existence should be noted (some paleontolo-

gists even doubted that such transitional forms ever lived, see Chap-

ter 14).

From the humble beginning in the cotylosaurs, the reptiles blos-

somed out during the subsequent geological periods in a variety of

creatures ranging in size from a mouse to giants with bodies up to 87

feet long, some living on land and others in water or in the air. When

a group of organisms becomes diversified in time into subgroups with

different body structures and ways of life, it is said to undergo adaptive

radiation. The following very brief account shows how great was the

adaptive radiation of the reptilian stock.

By far the most famous of the fossil reptiles were the dinosaurs,

some of which (though by no means all) reached enormous sizes.

They belonged to two distinct orders, one of which developed a struc-

ture of the pelvis bones resembling that found in birds, although these

dinosaurs were not the ancestors of birds. The similarity of the pelvis

structure is, then, an instance of convergent evolution. Some of the

dinosaurs were flesh eaters, preying probably on other reptiles. Ty-

ranosaurus stood some 19 feet tall, which makes this Cretaceous dino-

saur the largest known carnivore. The largest dinosaurs (Diplodocus

and Brachiosaurus, the latter estimated to have weighed some 50 tons)

were, however, vegetarians. It is hard to see how animals of such

bulk could have moved on land, and some paleontologists conjecture

that they waded partly submerged in shallow swamps or lakes. Stego-

s.turns and Ankylosaurus had bodies armored with bony plates, and

were slow-moving herbivores, perhaps not unlike the living hippo-

potami. Triceratops had two horns above the eyes, and one on the

nose, like a rhinoceros.

Other reptiles invaded the air, and produced the order of ptero-

saurs. These are again not the ancestors of the birds. Their flying

292 Historical Record of Organic Evolution

apparatus consisted of wings formed apparently by a skin fold like the

wings of modern bats, but supported only by a greatly enlarged fourth

finger. The pterosaur wing is, then, completely homologous neither

to a bird wing nor to a bat wing, although it performed the same

function, that is, flying. At least two reptilian orders, ichthyosaurs

and plesiosaurs, reverted to life in water, and transformed their walk-

ing legs into paddles or fin-like structures, although they, of course,

continued to respire by lungs. Some ichthyosaurs probably looked

superficially like the modern dolphins and porpoises, and led similar

lives.

Just what caused the dying out of these varied and apparently well-

adapted animals at the end of the Cretaceous period is obscure. The

mammals at that time were undergoing an adaptive radiation of their

own, and soon, together with the birds, replaced the reptiles. How-

ever, it is doubtful that the previously dominant reptiles could have

been destroyed by the mammals. More probably the reverse was

true: the removal of the reptiles opened the evolutionary opportuni-

ties to the mammals and birds, and they responded by adaptive radia-

tions, which filled the "vacancies" in the economy of nature hereto-

fore preempted by the reptiles.

It is also instructive to note that the ancestors of the now-living

reptiles played rather inconspicuous roles during the Age of Reptiles.

Lizards and snakes are the most successful of the modern reptiles,

yet lizards have been known only since the Jurassic period, and snakes

since the Cretaceous; and these orders can in no sense be regarded as

replacing the extinct "ruling" reptiles—in the biological domains of

the latter. Two other orders of reptiles, the crocodiles and the turtles,

exemplify a different situation. Crocodiles have been known since

the Triassic, and the turtles perhaps even since the Permian periods.

Both orders were more abundant and diversified in the past than they

are now, but by and large they underwent no major evolutionary

changes. Although the dinosaurs and other orders had their time of

rapid evolution, triumphant success, and then downfall and complete

extinction, the turtles and crocodiles kept on placidly exploiting their

peculiar ways of life without striking success but also without irrevo-

cable disaster. They show an evolutionary conservatism, often met

with in organisms specialized for some peculiar ways of life in which

they have no serious competitors.

On an island off New Zealand, now carefully protected by the

New Zealand government, there lives a superficially lizard-like rep-

tile Sphenodon (or tuatara), which is a representative of a primitive

The Age of Mammals 293

reptilian order. This order may be a slightly modified descendant of

the reptiles which gave rise to dinosaurs, true lizards, and perhaps

other orders. Anyway, the living Sphenodon is not easily distinguish-

able from a fossil discovered in Jurassic rocks of an estimated age

of more than 135 million years. This is a case of evolutionary con-

servatism, amounting to a complete stagnation, and bringing the ani-

mal to the very brink of extinction. The fact that Sphenodon has

been preserved only in New Zealand will be commented upon below

(page 296).

The Age of Mammals. This name is applied to the time span from

the early Tertiary to the present, some 70 million years. During this

time the earth came to be dominated by the two highest classes of

backboned animals, mammals and birds, and also by the flowering

plants and by insects (some people prefer to call it the Age of

Insects).

Mammals and birds have reached a level of organization greatly

superior to that of the reptiles and other animals (Figure 12.2). The

most important advance, especially in the mammals, has to do with

improved care of the young. Lower animals produce generally nu-

merous offspring, most of which fall prey to environmental accidents

and to predators. An individual matters little; the gamble is that a

small part of the offspring will escape destruction. One of the trends

observed in the progressive evolution of the living world has been

to make the individual's survival not a matter of lucky chance but

of his constitutional and genotypic fitness to survive (see Chapter 14).

The progeny becomes, therefore, less numerous, but every infant is

provided with maximum protection against the hostile world. Mam-

mals are viviparous, the embryo developing inside the mother's body

(except the Australian duckbill and spiny anteater, which lay reptile-

like eggs). The great mammalian invention is the placenta, a nexus

of blood vessels of the embryo and of the wall of the mother's uterus,

through which the embryo is supplied with food and with oxygen

(the marsupials, of which the opossums and kangaroos are representa-

tives, have, however, no placenta or an imperfect one). Furthermore,

all mammals, including the duckbill and the spiny anteater, nurse their

young on milk secreted by mammary glands, which may be modified

sweat glands.

Another evolutionary invention, made by birds as well as by mam-

mals, is the warm-bloodedness, or, more precisely, a physiological

horneostatic (see page 13) mechanism which maintains the body

temperature constant. This mechanism permits the mammals and

294 Historical Record of Organic Evolution

Reptile

Small brain case

Simple

cheek teeth

Mommol

Expanded brain case

Several bones

forming Jaw

Stapes

Generally

small ilium

Continuing

through life

Variable

Skull and

teeth

Bauria

Opossum

Ear ossicles

Pelvis

Growth

Temperature

Complex

cheek teeth

Forwardly

extended

ilium

Limited

Fairly constant

Scales or skin

Body Covering

Reproduction

Eggs

No care of young

No milk

Live birth

and care

of young

Milk

Figure 12.2. Some of the structural and physiological differences distinguishing

reptiles (left) and mammals (right). (From Colbert.)

The Age of Mammals 295

birds to remain active regardless of the temperature fluctuations in

the environment. It is easy to see how great an advantage in the

struggle for existence is such an independence of the organisms from

the vagaries of weather. The development of hair (in mammals) and

of feathers (in birds) in place of the reptilian scales was probably a

concomitant of the regulation of the body temperature.

Mammals have developed also a leg posture superior to that of the

reptiles. The legs of amphibians and reptiles, having evolved from

the fins of their aquatic ancestors, are usually widely sprawled out

sidewise from the body, so that at rest the belly lies on the ground.

In mammals the legs have been placed under the body, which now

rests on the legs. This change, minor though it may seem, has neces-

sitated widespread alterations of the skeleton and the musculature.

Finally, mammals have greatly elaborated their teeth. Instead of the

simple peg-like teeth of the reptile, there is a great variety of mam-

malian teeth adapted for nipping, grasping, piercing, cutting, pound-

ing, or grinding food, the same jaw having different teeth in front and

on the sides. To be sure, mammals have sacrificed the indefinite

amount of tooth replacement; barring artificial ones, mammals have

just two sets of teeth, the "milk" and the "permanent" dentition (Fig-

ure 12.2).

A remarkable thing about the evolutionary history of mammals is

that they appeared long before they became dominant and the Age

of Mammals began. In fact mammal-like reptiles, the so-called therap-

sids, had appeared already at the beginning of the Age of Reptiles,

in the Triassic and even in the late Permian periods. The mammalian-

like characters of the therapsids are found in the structure of their

skeletons, since, obviously, there is no way of telling whether they had

already embarked on the development of the temperature regulation

and the improved care of the young. In the Jurassic period four differ-

ent orders of mammals already existed, none of which survives today.

They were small, rare, and inconspicuous animals. Unfortunately,

very few of them were preserved as fossils, and those that were are

represented not by complete skeletons but by fragments, mostly by

teeth. There is, again, no way of telling how rapid was the progress

in the evolution of mammalian physiological adaptations, since only

bone fragments are found. But by the late Cretaceous, as the Age of

Reptiles was drawing to its close, some mammals appeared whose

bones are indistinguishable from those living now, particularly opos-

sums and small insectivores not very different from modern shrews.

And only in the early Tertiary, during Paleocene and Eocene, did the

296 Historical Record of Organic Evolution

adaptive radiation of mammals burst forth, and every mammalian

order now living, and some that are extinct, became represented by

more or less numerous forms. (This radiation appears, however, less

sudden if we recall that the Paleocene and Eocene together lasted for

some 35 million years!) As indicated above, the coincidence of the

adaptive radiation in mammals with the disappearance of the previ-

ously dominant reptiles can hardly be due to chance, but the causal

relations between the two events are far from clear.

Evolutionary Rates. It should already be clear from the foregoing

discussion that the evolutionary changes do not proceed at anything

like uniform rates in all organisms. Organic evolution is in this re-

spect radically different from such inexorably even-paced processes

as the decomposition of radioactive elements. The evolution of some

groups appears to have come to a halt (Figure 12.3). The opossum,

which in late Cretaceous was one of the evolutionally most progres-

sive forms, has changed little duiing the 75 or more million years since.

The Sphenodon (see page 292) has not changed appreciably for some

135 million years, since the Jurassic. A "living fossil" which became

famous in 1939 is a coelacanth fish (Latimeria) which belongs to a

group believed to have been extinct since the Cretaceous period until

a fisherman caught it alive in the Indian Ocean, off South Africa. In-

cidentally, the coelacanths are regarded as the descendants of the

order of fishes which probably gave rise (during the Paleozoic time)

to the land vertebrates, and consequently to ourselves. The horseshoe

"crab" (Limulus), which is a familiar sight on the Atlantic Coast

beaches in late spring and early summer, is not very different from

some fossils which were alive some 200 million years ago. But the

animal that can be used as a symbol of a really staunch conservatism

is the little marine brachiopod Lingula, which has not changed, at

least on the outside, for about 400 million years, since the Ordovician

times (Figure 12.3).

At the opposite extreme are the dangerous radicals, which change so

rapidly that the fossil record becomes quite inadequate to register the

changes. The class of mammals as a whole belongs here, since, with

few exceptions indicated above, none of the modern forms existed

in the late Cretaceous. Few of the modern species of mammals ex-

isted even in the late Tertiary (although all the modern orders and

many families and genera did). The human species is one of the worst

radicals, since it evolved so rapidly that our ancestors only one million

years ago were quite appreciably different from ourselves.

Evolutionary progressiveness or conservatism are not permanent

298 Historical Record of Organic Evolution

properties of a group of organisms. Even Lingula must have at some

time evolved from more primitive ancestors. A rapidly evolving group

may enter upon a period of evolutionary stagnation, and the opposite

change may also occur. Some biologists supposed that every group

of organisms has its "evolutionary youth" when it evolves rapidly, its

maturity when it is common and produces many species and varieties

in different parts of the world and in different habitats, and its "evolu-

tionary old age," when it stops evolving or evolves in wrong direc-

tions, leading to extinction. This matter will be discussed in the con-

cluding chapter of this book. For the time being it is sufficient to

say that evolutionary "youth" and "senility" are words which are mis-

leading, since they suggest a wrong analogy between the develop-

ment of an individual and the evolution of a species, a genus, or a class.

The evolutionary fate is to a great extent determined by the relation-

ships between the organism and its environment, and has no inexor-

able course, as the life of an individual does. This fact can be seen

best by considering a concrete evolutionary history, and we choose for

this purpose the example of the evolution of the horse family (Equi-

dae), which has been studied in detail by some of the most eminent

paleontologists.

Evolution of Horses. According to Simpson, the foremost authority

on mammalian origins: "It may be, indeed, that the lion and the lamb

are cousins." He has said this because the ungulate mammals, to which

horses and lambs belong, and the carnivores, to which lions belong,

are greatly modified descendants of the closely related condylarths

and creodonts. Condylarths and creodonts were rather common ani-

mals during the earliest Tertiary epoch, Paleocene. They became ex-

tinct as such during the next epoch, Eocene, presumably being re-

placed by their own modified or better fit to survive descendants.

The best-known condylarth, called Phenacodus, resembled a horse

about as little as it did a lion or a lamb. It was an animal rather

smaller than the smallest pony, with feet having five toes, each toe

provided with a small hoof (Figure 12.4). Its teeth were small and

pointed, the canines being somewhat larger and remotely dog-like.

Such teeth suggest that Phenacodus was probably omnivorous, eating

the various foods, both vegetable and animal, that it could get.

During the Eocene, Hyracotherium lived. It is more widely known

under the name of Eohippus, the "dawn horse." There is a break in

the fossil record between the condylarths and Eohippus, that is, no

intermediates between them have yet been found. It is chiefly for

this reason that Eohippus is placed in the order of ungulates and in

302 Historical Record of Organic Evolution

12.6). The head of Eohippus was, again, not very different from the

head of Phenacodus, but the molar teeth had already begun to develop

the crests which serve to grind the food, whereas the front teeth (in-

cisors) became modified for nipping and picking up leaves, and a gap

(diastema) appeared between the incisors and the canines. This gap

is present in the later and in the modern horses. Eohippus did not

graze on grass, but probably browsed on leaves, buds, and perhaps

on fruits and seeds. Having the benefit of hindsight, we recognize

that Eohippus was evolving not in the direction of lions, but in the

direction of lambs, or rather of horses.

Many textbooks and popular accounts of biology represent the evo-

lution of the horse family as starting with Eohippus, and progressing

in a direct line towards the modern horse, Equus. This evolutionary

progress involved, allegedly, the animals getting steadily larger and

larger, while their feet were losing toe after toe, until just a single hoof

was left. According to Simpson, this oversimplification really amounts

to a falsification. In reality things happened in a far more complex,

yet more meaningful way, summarized in Figure 12.7, which still

shows only the principal events of a long history.

During the Eocene epoch Eohippus evolved slowly, and produced

several species and related genera (Orohippus, Epihippus) which lived

both in North America and in Eurasia. Later on, all the Old World

species died out, while one or more American species gave rise to a

new genus, called Mesohippus. During the Oligocene epoch, Meso-

hippus became a very common animal in North America to judge by

the large number of its remains that have been discovered. It was

taller than Eohippus (24 inches tall on the average), and it had long

and slender legs, each with three toes. The brain casts of Mesohippus

show another important advance—an increased size of the cerebral

hemispheres. The success of this animal may have been due as much

to its growing more intelligent as to growing more fleet-footed.

During the next epoch, the Miocene, the Mesohippus stock blos-

somed out in a variety of new genera, among which Anchitherium,

Hypohippus, and Merychippus are most important (see Figures 12.6

and 12.7). The first two were great travelers, and spread out from

their native North America to the Old World, where they became

quite common. These successful colonizers were getting somewhat

taller as time went on, but they still had three toes on each foot, and

their teeth indicate that they were, like their ancestors, browsing on

foliage. Merychippus stayed at home in North America, but it made

an evolutionary invention which affected all subsequent evolution of

Evolution of Horses

303

Figure 12.7. A schematic representation of the evolution of the horse tribe,

showing the geographic distribution of the different forms, and their mode of

securing food by browsing or by grazing. (From G. G. Simpson, by permission

of the Oxford University Press.)

304 Historical Record of Organic Evolution

the horse family. This invention is the high-crowned teeth with a

complex pattern of crests and ridges; the horses now living have teeth

of this kind, though, of course, they are much larger than those in

Merychippus. Such teeth made possible a change in diet: Merychip-

pus and its numerous descendants, including the modern horses, fed

by grazing on grass instead of by browsing on leaves.

It was not a fortuitous chance that the switch from browsing to

grazing took place in horse evolution during Miocene. In Eocene and

during much of Oligocene, most of the earth enjoyed warm and humid

climates. Tropical evergreen and temperate broad-leaved forests

were the widespread types of vegetation, and they offered ample food

to browsing horses. But as time went on, the climates tended to be-

come cooler, and in many parts of the world also drier. Grassy steppes,

prairies, and savannas were becoming more and more widespread.

And yet, during Miocene, few animals, or at least few mammals, were

adapted to utilize grass as their main diet. Grass is very harsh food,

and the low-crowned teeth of the browsing horses would have been

worn down to the gums by tough grass. Merychippus evolved teeth

which enabled it to feed on grass, and its descendants "inherited the

earth," or at least the grass-covered part of it. But the new way of

life necessitated also qualities other than high-crowned teeth. On

grassy plains, animals can be seen from greater distances than in for-

ests. Accordingly, natural selection favored larger, stronger, and

faster grazing horses, which could defend themselves or could escape

from their enemies.

The net result of the change in the diet was a quickening of the

pace of evolution in the Merychippus line. The horses of the next

epoch, Pliocene, were derived chiefly from Merychippus, whereas the

browsing horses lingered on, and eventually became extinct (Figures

12.5 to 12.7). One of the most successful descendants of Merychippus

was the genus Hipparion, which repeated the exploits of the Miocene

browsing horses, and spread from North America to the Old World,

colonizing not only Asia and Europe but Africa as well. This was an

animal of about the size of a modern cow pony, but it still had three

toes on each foot. Meanwhile, in North America there arose still

another horse, Pliohippus, in which the side toes were finally lost,

leaving just a single toe on each foot. Pliohippus appears to be the

progenitor of the genus Equus, to which belong the modern horses,

zebras, and asses (see Chapter 9).

The last chapter of the history of the horse family is a puzzling one.

During the Ice Age (Pleistocene) the genus Equus spread from North

Evolution and Opportunity 305

America to Eurasia, giving rise to species of horses and asses, to Africa

where zebras were formed, and finally to South America. Just what

the indigenous North American horses were like in color and in habits

is anybody's guess, but it is fairly certain that the early human in-

habitants of the Americas (Indians) found horses abundant in many

parts of the continents. It is quite certain, however, that by the time

white men came to the Americas, the native horses had died out, and

horses were re-introduced from Europe. What or who killed the

native American horses? Was it an infectious disease, or animal pred-

ators, or hunting by the Indians? The cause is unknown, and serious

objections can be raised against all these hypotheses.

Another puzzle of the horse evolution is that all major evolutionary

advances have taken place on the North American continent. As Fig-

ure 12.7 shows, horses repeatedly invaded the Old World from North

America, had a brief efflorescence there, but each time died out with-

out leaving descendants. Remains of Eohippus, Anchitherium, and

Hipparion were found in Europe before fossil horses were discovered

in America. The pioneer Russian paleontologist Kovalevsky, who

made a classical study of the European fossils in 1873, concluded that

Eohippus—Anchitherium—Hipparion—Equus formed a line of descent.

This was one of the first triumphs of evolutionism in paleontology.

Kovalevsky, of course, was right in his conclusion that the fossils he

was studying represented successive stages of horse evolution, but he

had no way of knowing that the forms which he had at his disposal

were blind offshoots of an evolutionary development which took place

on the other side of the globe. Only since the American paleontolo-

gists Marsh, Cope, Osborn, Matthew, and Simpson have discovered

and studied the American fossil horses has the evolution of the horse

family become one of the best-known and most instructive examples

of the historical development of a group of living beings.

Evolution and Opportunity. Although this interpretation is not

universal-ly accepted (see Chapter 14), it seems probable that the im-

portant advances in the evolution of the horse family occurred when

the animals adopted new ways of life. Thus the high-crowned teeth

with complex grinding surfaces arose presumably when the horses

were becoming grazers instead of browsers. In turn, the develop-

ment of the grazing habit was a response to the existence of an abun-

dant and hitherto little exploited food supply—grass. Such a causal

connection between ecological opportunity (a new and plentiful food

supply) and evolutionary change (development of organs able to uti-

lize this food) is exactly what is expected if evolution is brought about

306 Historical Record of Organic Evolution

by natural selection of fit gene combinations. Modern paleontologists

and biologists are certainly not yet in a position to give causal explana-

tions along similar lines to all the evolutionary changes of which the

fossil record bears witness. We can, however, take comfort in the

fact that as our knowledge both of living and of fossil organisms in-

creases, the ecological factors responsible for evolutionary changes

emerge more and more clearly in more and more evolutionary his-

tories.

One of the most fascinating and suggestive phenomena which can

be understood in the light of the above theory is evolutionary paral-

lelism. It has been pointed out in Chapter 10 that organisms which

are not closely related often develop analogous organs, serving similar

functions, when these organisms have similar ways of life. Thus

fishes, ichthyosaurs, and whales and porpoises evolved remarkably

similar streamlined body shapes. Comparison of the inhabitants of

certain countries reveals evolutionary parallelism on even grander

scale. We shall consider two examples in this category: comparison

of the mammals of Australia with those of the rest of the world, and

comparison of the histories of the mammals inhabiting North and

South America.

It has already been stated (page 293) that the mammals which give

birth to living young can be divided into placentals and marsupials.

In the former, the fetus obtains its nourishment from the mother's

body through the placenta, and is born at an advanced stage of de-

velopment. In the marsupials the young are born quite small, and are

taken care of in a special pouch on the mother's belly, which serves

as an incubator and a perambulator combined. In the world at large,

placental mammals greatly outnumber the marsupials. Thus in North

America the only marsupials are the opossums; in Europe there are

none at all. Australia, however, is an exception. The most conspicu-

ous and characteristic Australian mammals are marsupials. Further-

more, and this is the point worth a particular emphasis, among the

Australian marsupials many kinds of animals exist which resemble,

both in appearance and in ways of life, placental types living else-

where.

Figure 12.8 shows some examples. There exist wolf-like marsupials

(Tasmanian "wolf), mole-like, squirrel-like (phalangers), rodent-like

(wombats corresponding to our woodchucks) and even anteater-like

marsupials. There is nothing quite like horses, deer, antelopes, or

other ungulates in Australia, but their place is taken by the herbivorous

kangaroos, which are represented by many species of different sizes.

308 Historical Record of Organic Evolution

do not happen to colonize an island, are quite common (see page 313).

Australia offers about as rich a variety of ecological opportunities as

larger continents. It has temperate and tropical forests, steppes, des-

erts, mountains, and plains. To the existence of these opportunities

life has responded by evolving parallel adaptive organizations. A mar-

supial "mole" is astonishingly like a "real" placental mole, both in

general appearance and in habits. And yet any competent anatomist

sees easily that the similarities are due to analogy and not to homology

(see page 231). But the opportunity to develop a marsupial "bat" was

cut off by the invasion of placental bats from Asia; the bats can nego-

tiate marine straits perhaps more easily than non-flying mammals can.

The Mammalian Inhabitants of North and South America. The

drawback to the story of the Australian mammalian fauna is that its

geological history is little known. In this respect a comparison of the

mammals of North and South America is more instructive. For South

America was an island-continent on and off during its history, and

again it was repeatedly connected to North America by land bridges,

such as the Isthmus of Panama is at present. It happens that during

most of the Tertiary Age of Mammals, South America was an island.

On the contrary, North America, at least during the Tertiary, was

always connected with, or separated by only a narrow strait like the

present Bering Strait, from the great Eurasian land mass. The con-

nection was through Alaska and northeastern Siberia, and organisms

able to negotiate the climatic conditions of that region passed freely

between the New and the Old Worlds. We have seen that ancient

horses made use of this route several times.

By Eocene times both placental and marsupial mammals were

spread widely over the surface of the earth. Their fossilized remains

have been discovered in both North and South America. Of course

they belonged not only to extinct species, but also with few excep-

tions (such as the opossum) to families and even to orders which have

since become replaced by greatly modified descendants. From Eocene

onward, however, South America became an island, since there was a

sea strait in place of the Isthmus of Panama. The mammals of South

America evolved, then, independently from those of North America,

with little or no interchange of inhabitants between the two conti-

nents. The results of these independent developments were remark-

able.

During the later part of the Tertiary period (Miocene and Pliocene

epochs) both North and South America had diversified mammalian

Mammalian Inhabitants of North and South America 309

inhabitants, more than twenty different families having been preserved

as fossils on each continent (see Table 12.2). The striking fact, how-

TABLE 12.2

NUMBERS OF DIFFERENT FAMILIES OF MAMMALS WHICH ARE KNOWN TO

HAVE LIVED ON THE NORTH AMERICAN AND THE SOUTH AMERICAN CONTINENTS

AT DIFFERENT TIMES, INCLUDING THE FAMILIES WHICH WERE COMMON TO

BOTH CONTINENTS

(According to Simpson.)

Families in Families in Families in

Time North America South America Both Continents

Recent 28 30 15

Pleistocene 32 36 21

Late Pliocene 27 26 5

Early Pliocene 28 26 2

Late Miocene 26 24 1

Mid-Miocene 27 23 0

ever, is that almost all the North American mammals were quite dif-

ferent from the South American ones. Only a few families, and no

genera, lived on both continents. In other words, the independent

evolutionary developments have produced different animals in the

two Americas.

And yet, the North American and the South American Tertiary

mammals often represented strikingly parallel biological types. Thus

in North America (and in the Old World) there appeared dog-like

and cat-like carnivores, the order Carnivora belonging, of course, to

the placental mammals. In South America there also appeared some

animals resembling wolves and others resembling saber-toothed tigers.

But these South American carnivores are marsupials, not placentals!

In the Old World (including North America) herbivorous ungulates

developed, including horses, camels, rhinoceroses, etc. In South Amer-

ica two now completely extinct orders evolved, known as litopterns

and notoungulates, which have produced some forms amazingly simi-

lar to ungulates (Figure 12.9). Thus the family Proterotheriidae, be-

longing to the Litopterna, became horse-like, including the reduction

of the number of toes in one line to three and in the other to a single

toe on each foot. These one-toed litopterns had already appeared in

Miocene, whereas one-toed horses are known only from Pliocene on.

Curiously enough, the three-toed and one-toed litopterns did not de-

velop grazing teeth; but another group (Notohippidae) did acquire

grazing teeth but not the legs specialized for running that horses have.

Oceanic and Continental Islands 311

to preempt the ecological niche, the profession, of a wolf-like animal.

Most of the peculiar and fascinating South American mammals of

the Tertiary have become extinct. In late Pliocene or in Pleistocene

times the Isthmus of Panama rose from the sea and made a land corri-

dor linking the two continents. Through this corridor the North

American mammalian families, including horses, tapirs, deer, camels,

cats, wolves, rodents, and many others, invaded South America. Simi-

larly, North America acquired armadillos, porcupines, sloths,' and other

forms from South America. During Pleistocene and Recent times

more than half of the mammalian families occurring on either conti-

nent occur also on the other continent (Table 12.2).

The invasion from the North had disastrous effects on many native

animals of South America: several whole orders and many families

died out. The litoptern pseudo-horse and the marsupial pseudo-wolf

were simply no match for the real horse and the real wolf. Strangely

enough, the immigrants from the South did not wreak any havoc in

North America, and the extinction of only a few North American na-

tives is ascribable to the competition of the invaders from the South.

Just why this should be so is a matter of speculation. One guess, put

forward originally by Matthew, is that, among land organisms, the

successful products of evolution are most likely to arise close to the

center of large continental masses, where the numbers of competing

forms are largest and the competition is toughest. Isolated islands,

and remote corners of continents, on the contrary, may harbor some

organisms relatively less resistant to the rigors of competition. During

the Tertiary, South America was an island while North America was

exchanging inhabitants with the great Asian-European-African land

mass. Many of the native South American mammals were, then, sup-

planted by the tougher invaders.

Oceanic and Continental Islands and Their Inhabitants. In 1835

Charles Darwin, then a young naturalist, visited the Galapagos Islands,

located on the Equator some 600 miles off the West Coast of South

America (Figure 12.12). He was greatly impressed by finding that

most of the animals and plants in Galapagos were endemic, that is,

occurred nowhere else in the world. At the same time many of them

resembled the inhabitants of the neighboring South America. Darwin

correctly surmised that the Galapagos endemics were more or less

profoundly modified descendants of immigrants from the South Ameri-

can continent. Here, then, the occurrence of an evolutionary process

was reasonably obvious. Some two decades later, A. R. Wallace, the

co-discoverer with Darwin of the principle of natural selection, reached

312 Historical Record of Organic Evolution

similar conclusions by studying the inhabitants of islands on the oppo-

site side of the globe—in Indonesia.

Two kinds of islands may be conveniently distinguished—the oceanic

(iii)

(iv)

(vi)

(viii)

(ix)

(xi)

(xii)

(X)

Figure 12.10. Darwin's finches of the Galapagos Islands. This is a group of

birds which became adapted to diverse modes of life and developed a great

variety of adaptations, particularly in the structure of the beak. (From Lack,

courtesy of Cambridge University Press.)

and the continental ones. Oceanic islands have never been connected

by land with any continent, or were so connected only long ago.

Galapagos and the Hawaiian islands are good examples of the oceanic

kind; they are islands of volcanic origin, and they rose from the bottom

of the ocean far from any land. Continental islands are separated

from the mainland generally by shallow straits, and were parts of the

mainland in geologically recent times. Long Island near New York,

Oceanic and Continental Islands 313

Newfoundland off the coast of Canada, or the British Isles off Europe

are examples.

Continental islands have rather few or no endemic plants or animals.

Their flora and fauna differ little from the adjacent mainland. The

flora or fauna of an oceanic island, on the contrary, are more or less

strongly different from those of other islands or continents. Not only

do oceanic islands have many endemics, but their floras and faunas

are usually "unbalanced," that is, they lack whole groups of plants

and animals which occur pretty much everywhere in the world where

suitable climatic and other conditions are available. Thus we know

that the oceanic island-continent of Australia has few placental mam-

mals. The oceanic Galapagos Islands have their mammalian inhab-

itants restricted to some bats and some rats, but they have instead a

remarkable fauna of reptiles, including the famous giant tortoises and

herbivorous lizards. Many kinds of birds familiar on the South

American continent are also missing on Galapagos, but one group of

birds, the finches, are represented by numerous and diversified forms.

These "Darwin's finches," recently subjected to penetrating study by

Lack, have evolved on Galapagos species with beaks adapted for

feeding on large or on small seeds, with strong woodpecker-like beaks

feeding on insect larvae, with parrot-like beaks feeding on buds and

fruits, with slender beaks feeding on small insects, etc.

. These facts are comprehensible only in the light of the history of

Iffe on oceanic islands. An island arising in the midst of an ocean

can be populated only through accidental introductions of plants and

animals by oceanic currents, by winds, storms, etc. Just which kinds

of plants and animals will arrive and when will depend, of course, on

their ability to remain alive during the transport. But it will depend

also on chance or luck; Simpson has called the accidental methods of

colonization of new territories sweepstakes dispersal. How low are

the chances of winning in these sweepstakes has been estimated by

Zimmerman for the Hawaiian Islands. These islands have about 3700

endemic species of insects and about 1000 endemic snails; Zimmerman

reckons that they may have arisen from about 250 insect and 25 snail

species introduced in Hawaii from elsewhere. On geological grounds

Hawaii is estimated to be of the order of 5 million years old. This

means a successful introduction of a new insect species once in 20,000

years, and of a snail species once in 200,000 years on the average!

The peculiar floras and faunas of oceanic islands are, therefore, due

to the accidental nature of their colonization. A finch, arriving in

Galapagos when these islands had few or no other bird inhabitants,

Zoogeographic and Phytogeographic Regions 315

matter may respond by evolution of forms adapted to occupy these

ecological niches.

Zoogeographic and Phytogeographic Regions. The second half of

the eighteenth century saw the completion of the first phase of the

geographic exploration of the world. By that time all the continents

and the major islands had been discovered. At the same time, the

descriptive studies of the geographic distribution of living creatures

began to reveal that every land area and every section of the sea have

their own sets of animals and of plants. Much depends, of course,

on the climate, but this is far from the whole story. Parts of California,

of Mediterranean countries, of Australia, South Africa, and western

South America have rather similar climates, and yet the native in-

habitants of these countries are quite different. Tropical rainforests

occupy extensive territory in the basin of the Amazon in South Amer-

ica, the Congo Basin in Africa, and in southeastern Asia and Austral-

asia, but the plants and animals of these rainforests are different.

These facts were generalized during the second half of the nineteenth

century, when the world was subdivided into several zoogeographic

(faunal) and phytogeographic (floral) regions. Although the de-

tails of this subdivision are not even now fully agreed upon, the map

in Figure 12.12 summarizes the essentials of the story.

The United States, Canada, and the highlands of Mexico are in-

habited by the Nearctic fauna. This fauna is not very different from

the Palearctic fauna—in Europe, Asia down to the southern slope of

the Himalaya Mountains, and Africa north of Sahara Desert. The

Nearctic and Palearctic fauna may, consequently, be parts of a larger

Holarctic fauna. South America, Central America, and the Mexican

lowlands are inhabited by the Neotropic fauna. Africa south of

Sahara has the Ethiopian fauna, southern Asia and Indonesia the

Oriental or Indo-Malayan fauna; and Australia is a world unto itself—

here is the Australian fauna.

Darwin, Wallace, and among their successors particularly Lydekker

(1896) and the Sclaters saw clearly that the existence of this regional

differentiation of the living world is comprehensible only as an out-

come of an interaction between the geological history of the earth and

the organic evolution. We have already discussed the history of the

Australian region (page 307). The marsupial mammals underwent

there an adaptive radiation, being protected from competition of the

placental mammals by the island nature of Australia. The Australian

tree flora is dominated by species of eucalypts and of acacias, which

have evolved almost every conceivable ecological type of tree. Out-

Suggestions for Further Reading 317

arose in connection with the hypothesis of continent drift developed by

the German geologist Wegener. According to this hypothesis, conti-

nents are blocks of relatively light rocks floating on the surface of the

viscous and heavier strata of the earth's crust. Wegener visualized

that the present continents during the Paleozoic and earlier times

formed one or two enormous land masses. Africa was united with

southern Asia, and through South America also with Antarctica and

Australia, into a single mass. This mass broke up into the present

continents, which then proceeded to drift apart, the Atlantic, Indian,

and Antarctic oceans forming between them.

Some zoologists and botanists believed that the continental drift

hypothesis explained some faunal and floristic resemblances, particu-

larly between Africa and South America, South America and Australia,

and Africa and Australia. Indeed, there are examples of animals and

plants which occur on almost any pair of continents now separated by

wide oceans, and not in the territories linking these continents. In

Chapter 9 we saw a situation of this sort exemplified by species of

cottons. However, the hypothesis of continental drift has encountered

grave and apparently insuperable difficulties, and is now abandoned

by most geologists. The biological facts which were adduced in its

favor were never compelling, and most of the modern biogeographers

believe that this hypothesis should be rejected also on biological

grounds. The scattered instances of discontinuous distribution of

animals and plants on remote continents have to be considered, each

on its own merits. Sometimes we are dealing with organisms which

were widely distributed in the past, but now survive only in some

widely separated territories. Thus species of incense cedar (Libo-

cedrus) are now known in California, Chile, and New Zealand. Some-

times there are instances of "sweepstakes distribution."

Suggestions for Further Reading

Simpson, G. G. 1944. Tempo and Mode in Evolution. 1st Edition. 1953. The

Major Features of Evolution. 2nd Edition. Columbia University Press, New

York.

Simpson, G. G. 1949. The Meaning of Evolution. Yale University Press, New

Haven.

Many paleontologists questioned whether the evidence of the fossil record could

be reconciled with the biological theory of evolution. Simpson's boote have shown

that a synthesis of paleontology and the rest of evolutionary biology is, indeed,

possible. In fact, these works have to a considerable extent created the modern

version of this theory- They should be read by every evolutionist.

318 Historical Record of Organic Evolution

Dunbar, C. O. 1949. Historical Geology. John Wiley, New York.

Colbert, E. H. 1951. The Dinosaur Book. 2nd Edition. McGraw-Hill, New

York.

Colbert, E. H. 1955. Evolution of the Vertebrates. John Wiley, New York.

Gregory, W. K. 1951. Evolution Emerging. Macmillan, New York.

Romer, A. S. 1941. Man and the Vertebrates. University of Chicago Press, Chi-

cago.

Romer, A. S. 1945. Vertebrate Paleontology. 2nd Edition. University of Chi-

cago Press, Chicago.

Simpson, G. G. 1950. History of the fauna of Latin America. American Scien-

tist, Volume 38, pages 361—389. Also in Baitsell's Science in Progress, Volume 7.

Simpson, G. G. 1951. Horses. Oxford University Press, New York.

The basic facts of historical geology and paleontology can be found in the eight

books (Dunbar to Simpson) cited above.

Mayr, E. 1944. Wallace's line in the light of recent zoogeographic studies.

Quarterly Review of Biology, Volume 19, pages 1-14.

Mayr, E. 1946. History of the North-American bird fauna. Wilson Bulletin,

Volume 58, pages 3—41.

Matthew, W. D. 1939. Climate and Evolution. 2nd Edition. New York Acad-

emy of Science, New York.

Simpson, G. G. 1953. Evolution and Geography. Oregon State System of

Higher Education, Eugene.

Stebbins, G. L. 1950. Variation and Evolution in Plants. Columbia University

Press, New York.

Chapter XIV contains a concise discussion of phytogeography. The works of

Mayr, Matthew, and Simpson discuss fundamentals of zoogeography.

The following three works are concerned specifically with the faunas of oceanic

islands:

Amadon, D. 1950. The Hawaiian honeycreepers. Bulletin of the American

Museum of Natural History, Volume 95.

Lack, D. 1947. Darwin's Finches. Cambridge University Press, London.

Zimmerman, E. C. 1948. Insects of Hawaii. University of Hawaii Press, Hono-

lulu.

13

Human Evolution

Darwin's discovery that man is a descendant of non-human ances-

tors seemed repugnant to some of his contemporaries. The story goes

that, on hearing about Darwin's theory, a lady cried: "Descended •

from the apes! My dear, we will hope that it is not true. But if it is,

let us pray that it may not become generally known." To her, it was

terribly degrading to be related, however distantly, to an ape. But

the news became rather generally known, and most people grew

reconciled to the strange relative.

At present the dust has settled sufficiently to see things more clearly.

Man is a biological species, subject to the action of biological forces,

and a product of a long evolutionary development. It does not matter

whether the evolutionary origin of man is called an "hypothesis" or

a "fact." Events which occurred before there were observers capable

of recording and of transmitting their observations must of necessity

be inferred from evidence now available for study. But the evidence

shows conclusively that man arose from forebears who were not men,

although we have only the most fragmentary information concerning

the stages through which the process has passed. Nobody has seen

that the earth is a sphere or that it revolves around the sun, rather

than vice versa; nobody has caught a glimpse of atoms or of things

within atoms. Are atoms, then, factual or hypothetical? The least

that can be said is that in our activities we take the earth to be a

sphere and treat atoms as though they were facts. For similar reasons,

it is not a matter of personal taste whether or not we "believe in"

evolution. The evidence for evolution is compelling. Moreover,

human evolution is going on at present, and, what is more, biology is

319

320

Human Evolution

in the process of acquiring knowledge which may permit man to con-

trol and to direct this evolution.

Biologists have been so preoccupied with proving that man is a

product of organic evolution that they have scarcely noticed that man

is an extraordinary and unique product of this evolution. He is unique

in a purely biological sense. Some forces which are important in the

evolution of man occur at most as vestiges in the evolution of other

creatures. The leading forces of human evolution are intelligence,

ability to use linguistic symbols, and the culture which man has de-

veloped. These exclusively, or nearly exclusively, human phenomena

affect the biological evolution of man so profoundly that it cannot be

understood without taking them into account. Conversely, human

society and culture are products of the biological evolution of our

species. Human evolution is wholly intelligible only as an outcome

of the interaction of biological and social forces. Biologists and so-

ciologists are equally guilty of underestimation of this fact.

Characteristics of the Class of Primates. Although Linnaeus was

not an evolutionist, he correctly placed man, Homo sapiens, in the order

of primates, of the class of mammals, of the vertebrate phylum.

Except for man, who now lives in all climates which the earth has to

offer, primates are tropical or subtropical animals. Several groups of

primates are known, as shown in Table 13.1 and Figures 13.1 and 13.2.

TABLE 13.1

SYSTEMATIC POSITION OF MAN AND His LIVING RELATIVES BELONGING TO

THE ORDER OF PRIMATES OF THE CLASS OF MAMMALS

Suborder

Anthropoidea

Prosimii

(After Simpson.)

Infraorder or Vernacular Name and Geographic

Super-family Distribution

Family Hominidae, Men (World-wide)

Hominoidea Family Pongidae, Apes (Tropical Asia

and Africa)

Cercopithecoidea Old World Monkeys (Tropical Asia and

Africa)

Ceboidea New World Monkeys (Tropical America)

Tarsiiformes Tarsiers (Indonesia, Philippines)

Lorisiformes Loris and Galagos (Tropical Asia and

Africa)

Daubentonioidea Aye-Aye (Madagascar)

Lemuroidea Lemurs (Madagascar)

Tupaioidea Tree-Shrews (Tropical Asia)

Living Relatives of Man 321

Very few distinctive traits are common to all primates. Most, though

not all, occur in tropical forests and are tree dwellers. The thumb,

and usually also the great toe, are opposable to the other digits—an

advantage for creatures that must climb and jump among tree

branches. Man, however, has lost the "opposability" of the great toe

which his ancestors possessed. Most, though again not all, primates

use their hands and feet for grasping and handling objects rather than

for walking. Man's ancestors lost the use of their feet for grasping as

they attained their erect posture. The emancipation of the hands

from walking duties made possible their use for delicate manual oper-

ations of which man alone is capable. The teeth of primates are not

specialized for just one kind of food, as the teeth are specialized in,

for example, grazing animals, in rodents, or in carnivores. Primate

teeth deal successfully with diversified diets, both vegetable and ani-

mal. In mammals the eyes are placed usually on the two sides of

the head, permitting a very broad field of vision (consider the posi-

tion of the eyes, for example, in a rabbit or in a horse). In most

primates both eyes are directed forward, as in man. This position

facilitates binocular vision and correct estimation of distances, which

is vitally important in animals which dwell among branches of tall

trees.

Living Relatives of Man. Man resembles the anthropoid apes (fam-

ily Pongidae, Table 13.2) in body structure more than any other group

of animals. This family comprises two species of gibbons, living in

southeastern Asia, orangutan on the islands of Sumatra and Borneo,

and chimpanzee and gorilla in equatorial Africa. Not only do the

anthropoid apes share with man a relatively large size and many de-

tails of the structure of the brain and the skull, but they approach each

other in posture, in position of internal organs, absence of a tail, and

in many physiological traits, especially those dealing with reproduc-

tion (menstruation, the position of the fetus in the womb, etc.). The

differences between the families Hominidae and Pongidae are those

which are expected between animals which walk erect on the ground

and animals which dwell among tree branches.

About 140 species of New World monkeys and some 200 species of

Old World monkeys have been named. The number of species is in

reality only a fraction of the above numbers, because of the habit of

some taxonomists of giving species names to races of the same species.

Nevertheless, a considerable diversity of forms is involved. Monkeys

vary in size from marmosets, which are about as small as squirrels,

to baboons, which are of the size of large dogs. Some have long and

324 Human Evolution

powerful prehensile tails which act as a "third hand," but others

(dwelling on the ground rather than in trees) are tailless.

The primate suborder Prosimii (Table 13.1) stands between the

anthropoids on one side and the non-primate mammals on the other.

Lemurs and lorises (Figure 13.1) vary in size from that of a cat down

to that of a newborn kitten. In contrast to monkeys and apes, they

have movable and usually large ears, and elongated snouts rather

than flattened "faces." Their tails are usually long and bushy but

not prehensile. Lemurs do not menstruate. True lemurs are a re-

lict group which is now confined to the island of Madagascar in the

Indian Ocean off Africa (see Figure 12.12). Their fossil remains,

however, occur over much of the Old World and North America. A

curious creature called aye-aye (Daubentonia) has remarkably large

front teeth which resemble those of mice and other rodents, and has

claws on all digits except the big toes, which have flat nails. The

third finger of each hand is much longer than the others; the animal

uses this long finger to extract insects, on which it feeds, from cracks

in the bark and similar places.

The spectral tarsier (Tarsius spectrum, Figure 13.1) is the only

living representative of a group of primates which was much more

common and widespread in the past, to judge from fossil remains de-

scribed from Eocene strata of North America and Europe. The struc-

tures of the face, lips, and brain of the spectral tarsier make this little

animal (about the size of a two-week-old kitten) a connecting link be-

tween the Prosimii on one side and the anthropoids on the other. And

yet the tarsier shows several features which make him quite unique.

His eyes are relatively larger than in any other primate (it is an ex-

clusively nocturnal animal), and his hind legs are modified to enable

him to make powerful jumps among tree branches. The tarsioids of

the past may have given rise to the lemuroid stock on one hand and

to the anthropoid stock on the other. The spectral tarsier is an aber-

rant and greatly modified relic of the primitive tarsioids known only

as fossils. Wood Jones has even conjectured that the human stock

arose directly from the tarsioids, rather than through monkey-like and

ape-like ancestors. This conjecture is far-fetched, and it is mentioned

only to emphasize the interest of the tarsioids as a group which may

have been ancestral to other groups of primates.

Even more remarkable than the tarsioids are the tree shrews (Table

13.1 and Figure 13.1). These animals have a superficial resemblance

to squirrels, both in size and in general aspect. But their body struc-

tures are such that some zoologists have placed them in the order of

Fossil Non-human Primates 325

insectivores (to which belong moles, ordinary shrews, and hedgehogs),

but other zoologists have placed them among the primates. Now the

insectivores are generally considered to be the most primitive order of

living placental mammals (see Chapter 12), whereas the primates, to

which we belong, are at or near the top. The tree shrews indicate,

then, the probable origin of the ancestral primates from ancient in-

sectivores. Tree shrews have claws instead of nails on their fingers

and toes, which give to their feet a non-primate-like appearance, but

their digits have a greater range of movements than those of mammals

other than primates. Their brain is too small for a primate but too

large for an insectivore, and it shows a considerable development of

the part of the cortex concerned with the function of vision.

Fossil Non-human Primates. The fossil record of primates is un-

fortunately meager. As stated above, most primates dwell in tropical

forests, an environment highly unfavorable for the preservation of the

remains of living creatures as fossils. Furthermore, whatever remains

are found are usually fragments of bones and teeth. It is a difficult

matter to get from such fragments an idea of what the whole animal

might have been like. Because of this difficulty the disagreements

between different authorities on fossil human and prehuman beings

are often extreme and rather exasperating to non-specialists. Never-

theless, so great is the interest in human evolution that the utmost

efforts are expended to find fossil evidence, and whatever is discovered

is studied in the minutest detail. It is not too optimistic to hope that

much more satisfactory fossil evidence, than is now available will be

forthcoming before very long.

During the Paleocene and Eocene (see Table 12.1), there existed

several families of animals, about rat-like or mouse-like in size, which

had an unmistakable resemblance to modern tree shrews. During the

Eocene, Europe and North America enjoyed tropical climates, and

among their inhabitants were some tarsioids and lemurs. By the

Oligocene and Miocene, the prosimians disappeared in Europe and

North America, which became climatically unsuitable. Traces of

them, however, have been found in the Miocene of eastern Africa,

and in the Pliocene abundant remains were preserved on the island

of Madagascar, where living lemurs are found now.

The most ancient known ape is represented by a fragment of a

mandible found in the lower Oligocene strata of Egypt to which the

name Parapithecus has been given. In some respects this mandible

resembles the tarsier, which agrees with the notion that the Anthro-

poidea arose from a tarsioid stock (Table 13.1). By the Miocene a

326 Human Evolution

variety of ape-like animals existed in Africa and in Asia. A number

of fragmentary remains have been dug up in Kenya (east-central

Africa); a small form, called Limnopithecus, may have resembled the

gibbon (which now lives only in southeastern Asia). A larger one,

Proconsul, discovered in 1948 by Leakey, may have been an ancestor

or a relative of the living chimpanzee, but it had certain human fea-

tures which are absent in the chimpanzee. A little later, by the mid-

dle and upper Miocene, there lived in Africa, India, and southwest Eu-

rope (Spain) the fossil apes Dryopithecus and Sivapithecus. They

show many similarities with the now living great apes (chimpanzee,

gorilla, orangutan), and yet some characters of their teeth adum-

brate those known in fossil men. Fragments of a thigh bone and of an

arm bone suggest that Dryopithecus did not have the long and power-

ful limbs which the modern great apes have developed to enable them

to live in the tropical forest trees. The human ancestors probably

were not as strongly specialized tree dwellers as the modern apes are.

Discovery of complete remains of Dryopithecus must be awaited be-

fore we can judge whether this animal can be regarded as one of the

ancestors of man.

Man-Apes of South Africa. It is clear that a large-scale adaptive

radiation of anthropoid primates took place during the Tertiary pe-

riod, particularly in Africa. Some of the most interesting products of

that radiation lived in the southern part of that continent. Exception-

ally abundant, if fragmentary, relics of the animals concerned, coming

from at least 65 individuals, have been discovered in recent years by

Dart, Broom, Robinson, and their collaborators. Counting isolated

teeth, the remains may belong to over 100 individuals. The rate of

discovery has been so high that unprecedentedly abundant material on

fossil man may soon be accumulated. Those fossils have been de-

scribed under several generic names (Australopithecus, Plesianthropus,

Paranthropus, etc.), although it is most probable that the animals of

which these fossils are fragmentary remains belonged to only one or

two closely related species of the same genus, Australopithecus.

No agreement has been reached concerning the position of Australo-

pithecus among the primates. Some authorities regard it as a peculiar

ape which had nothing to do with human descent. Others, and prob-

ably a majority, put it in a special subfamily, Australopithecinae, which

has no living representatives, but in which the characteristics of the

human family (Hominidae) and of the ape family (Pongidae) are

blended in such a way that it is arbitrary to say whether the Australo-

pithecinae are to be considered apes or hominoids.

328 Human Evolution

shells, Dart and others conclude that Australopithecus was "an ani-

mal-hunting, flesh-eating, shell-cracking, and bone-breaking ape." Cer-

tainly no living ape subsists on such a diet. The possibility that Aus-

tralopithecinae were ancestors of mankind is seriously considered by

some authorities. For the time being this is only a possibility. Most

unfortunately, the geological conditions in which the Australopithe-

cinae fossils were found make precise determination of their age

doubtful. They may extend in time from as early as the Upper Plio-

cene to as late as the Middle Pleistocene; in other words, they may

be appreciably older or appreciably younger than one million years.

If they lived well into the Pleistocene, it would follow that they were

contemporaries, rather than ancestors, of early man. Robinson be-

lieves that Australopithecus in fact coexisted with man.

Java and Peking Men. Early hominoids which possessed enough

intelligence to fashion and use stone tools and yet had some unmis-

takably ape-like features lived during the Pleistocene (Ice) Age in

Asia. In 1891 the Dutch anthropologist Dubois found the first re-

mains, a skull cap and a thigh bone, of Java Man, whom he called

Pithecanthropus erectus (ape-man walking erect). In 1936 and later,

von Konigswald and others found in the central part of the island of

Java the remains of two more Java men. Between 1929 and 1940,

Black and Weidenreich found in caves near Peking, China, the remains

of perhaps as many as forty individuals of Peking Man, which became

known as Sinanthropus pekinensis. These Latin names, however, are

misleading, since we are dealing undoubtedly not with distinct genera

and species but merely with races of a single species of man. A bet-

ter terminology is (according to Mayr):

Homo erectus erectus = Java Man = Pithecanthropus.

Homo erectus pekinensis = Peking Man = Sinanthropus pekinensis.

Java Man lived during the Pleistocene age (first Interglacial to

second Interglacial times), perhaps some half million or more years

ago (Figure 13.4). He was rather undersized by modern standards

(average height estimated at about 5 feet); and the brain size varied,

in the known skulls, from 750 to 900 cubic centimeters, compared to

900 to 1200 cubic centimeters in Peking Man, and an average of about

1350 cubic centimeters in modern man. The skull bones are consist-

ently thicker than in modern skulls, the foreheads slope gradually

backwards from very prominent eyebrow ridges which are continuous

above the root of the nose. The lower jaws are large and powerful,

with large teeth and without a chin. This must have made the face of

Homo erectus markedly prognathous, that is, the mandible, front teeth

330

Human Evolution

Neanderthal Men. Europe and North America offered some very

harsh environments during the Pleistocene Age. Four times enormous

ice sheets covered large parts of these continents, making climatic

conditions perhaps not unlike those now found in Greenland. The

four Ice Ages were separated by warmer interglacial times (Table

13.2). We are living during what may prove to be the Fourth Inter-

TABLE 13.2

RELATIVE CHRONOLOGY OF HUMAN PHYSICAL AND CULTURAL EVOLUTION DURING

THE ICE AGE

(After Vallois and Movius, Oakley, Le Gros Clark, and other sources.)

Fossil Human Races

Time and Species Cultures (in Europe)

Modern Races of sapiens Historic

Neolithic (New Stone

Age)

Mesolithic (Middle

Stone Age)

Cro-Magnon and relatives Magdalenian

Solutrean

Aurignacian-Perigordian

Mousterian |

Fourth Glacial (Wunn)

Third Interglacial

Third Glacial (Riss)

Second Interglacial

Second Glacial (Mindel)

First Interglacial

First Glacial (Gunz)

Neanderthal

Mount Carmel

Neanderthal

Fontechevade

Ehringsdorf

Steinheim

Peking Man(?)

Swanscombe

Peking Man

Java Man

Peking Man(?)

Java Man

Java Man, Australo-

pithecus(P)

Australopithecus( ?)

Levalloisian

Acheulian, Clactonian

Clactonian, Abbevillian

Abbevillian

Villafranchian

glacial Age. Concerning the duration of the Pleistocene Age and of

its glacial and interglacial stages, specialists are not in agreement.

One school (Milankovitch and Zeuner) thought that the Pleistocene

might have been about 600,000 years long, and that the Fourth

(Wiirm) glaciation lasted from about 125,000 to about 20,000 years

ago. Now it seems more probable (according to Movius, Flint, and

others) that the Pleistocene was perhaps one million years long. There

are as yet no definitive estimates of the dates of the various Glacial and

Emergence of Modern Man 331

Interglacial ages, and therefore we are giving in Table 13.2 what is

known as "relative" chronology of various findings of fossil man, rather

than "absolute" chronology (in years).

In any case, during the Third Interglacial and the Fourth Glacial

ages, Europe, western Asia, and north Africa were inhabited by a very

distinctive Neanderthal race of man (Homo sapiens neanderthalensis),

of which the remains of close to 100 individuals in varying degrees of

preservation have been found in places ranging from the Atlantic

coast of Europe to Asia (France, Gibraltar, Italy, Germany, Yugo-

slavia, southern Russia, Turkestan, Palestine).

The Neanderthalians (Figure 13.4) were of short stature, about 5

feet, but of exceedingly rugged build. They had a stooping posture,

powerful neck muscles, and massive heads with surprisingly large

brains—1450 cubic centimeters on the average, that is, as large as or

larger than in modern man. The bones of the skull were thick; the

brow ridges were large, although not as prominent as in Java Man.

The forehead was retreating, and the braincase flattened. The lower

jaw was heavy, without a chin, but with rather large teeth, and with

attachments for strong muscles which operated the jaws.

The Neanderthal race attained a relatively high cultural level. Al-

though the stone tools which they used were primitive, belonging to

the Old Stone Age types, the Neanderthalians had a social organiza-

tion which enabled them to hunt large and powerful game such as

mammoths and the woolly rhinoceros which at that time inhabited

Europe. There are evidences of religious activities, since their bones

are found under conditions suggesting ceremonial burial. In a Nean-

derthalian cave in Switzerland an altar has been found on which a

bear skull was placed.

Emergence of Modern Man. Some 75,000 years ago, while Europe

was in the grip of the last Ice Age, the Mousterian culture of Neander-

thal man was replaced by the Aurignacian culture, and the Neander-

thalian inhabitants by another race, which, to judge from its bones,

was like the modern man—Homo sapiens sapiens. The Aurignacian

stone tools are more expertly made than the Mousterian ones; the

makers, named Cro-Magnons from the cave in which the first remains

were discovered, were fine physical specimens, up to 6 feet tall and

with brains up to 1650 cubic centimeters in volume (Figure 13.4).

There is, of course, no way to tell what they were like in such external

traits as skin color and hair shape. Some of the remains in southern

Europe yield measurements which can be matched in the skeletons

of certain living Negro populations in Africa, but it does not follow

332 Human Evolution

that these early Europeans were Negroes in appearance or, even less,

in culture. Anyway, from the Aurignacian-Perigordian times on,

Europe and probably the rest of the world as well, was inhabited by

Homo sapiens sapiens.

Where did modern man come from? Where did he first arise? The

problem is very complex, speculation concerning it rife, and no con-

vincing solution is yet in sight. Some students have conjectured that

modern man developed in Africa, others that he came to Europe from

Asia, and destroyed the Neanderthalian natives. According to some

authorities, Europe had some human inhabitants who resembled mod-

ern men during the second interglacial period (350,000 or more years

ago), before the appearance of the Neanderthalians. Very few re-

mains of this hypothetical race have been found, and those which have

come to light are fragments not easily interpretable. The skull frag-

ments found at Fontechevade in France, at Swanscombe in England

and at Steinheim and at Ehringsdorf in Germany may have been rather

sopt'ens-like, despite their great age. It looks as though there lived

in Europe during the middle of the Ice Age a race rather like our-

selves and yet not ancestral to us. Some experts, in a kind of despera-

tion, regarded all fossil hominoids, except the Cro-Magnons and more

recent races, as collateral branches of the human family tree, which

became extinct without contributing to the direct ancestry of modern

man. This makes the origin of modern man more puzzling than ever.

Light is shed on this problem by the work of McCown and Keith

(1939), who studied a fine series of human remains found in the

caves of Mount Carmel in Palestine. These people lived at about

the same time that Europe was inhabited by the Neanderthalians.

But the dwellers of Mount Carmel range in physical type all the way

from indubitable Neanderthalians to more sapiens-\ike individuals.

The most reasonable explanation of this is that Palestine formed, at

that age (probably the third Interglacial), the geographic boundary

between the Neanderthal race (Homo sapiens neanderthalensis),

which lived to the northwest, in Europe, and a more nearly modern

race (Homo sapiens sapient) which lived perhaps farther southward,

in Africa. The inhabitants of Palestine were, then, either hybrids be-

tween the two races, or else they were an intermediate population of

a kind usually found at race boundaries (see Chapter 7).

In either case, one thing that is certain is that the Neanderthalians

belonged to the same species as ours. The replacement of the Nean-

derthalians by the Cro-Magnons in Europe may have occurred partly

by destruction and partly by hybridization with the incoming race,

Specific Unity of Mankind 333

which possessed a superior technology. Coon and other anthropolo-

gists are probably right that some modern European populations show

persistence of genes derived from the Neanderthalians. In like fash-

ion, within the last few centuries the white race has displaced the

red-skinned men in the Americas and the blackfellows in Australia.

It is possible that the Neanderthalians themselves had, at a much

earlier time, displaced the Swanscombe and similar populations which

had inhabited Europe before them.

The cultural development of the Cro-Magnons and other related

races who soon appeared in Europe was superior to that of the Nean-

derthalians, which might explain the rapid replacement of the latter

by the former. The Cro-Magnons and their relatives fashioned more

efficient stone tools; they probably knew the use of the bow and of

harpoons. Most remarkable of all, they have left superb drawings

of animals on the walls of the caves which they inhabited (Figure 1.1).

The artistic feeling which these drawings display is admired even

by modern man.

Specific Unity of Mankind. It is important to keep in mind that

new species do not arise in any single place but in large territories

(except polyploid species, see Chapter 9). A species is a Mendelian

population which lives in a more or less extensive area and which grad-

ually alters its genetic composition. Students of fossil man have a

habit of giving resounding Latin specific and generic names to almost

every bone fragment which they discover, which conveys the mis-

taken impression that there existed in the past many different man-like

species and genera. Weidenreich, however, has pointed out that when

the remains of human or prehuman forms which lived more or less

simultaneously are compared, the differences between them are only

of the order of those found between the now living human races (see

above). There is no fossil evidence of the existence at any one time

of more than a single human or human-like species, except, possibly,

in the case of the Australopithecinae (see above). Mankind preserved

its specific unity throughout its evolutionary development during the

Pleistocene times, although it always was, as it still is, subdivided into

races. Human evolution never led to differentiation of a single species

into a group of derived species, some of which might have become

lost and others survived. Vlankind was and is a species which evolves

as a body.

Any living species, race, or population tends to expand in numbers

as soon as it encounters favorable environment. In doing so, a popu-

lation often overflows into neighboring territories occupied by other

334 Human Evolution

populations, and the immigrants usually intermarry with the natives.

If the genotype of the immigrants is adaptively superior to that of

the natives, the population resulting from the mixture comes to re-

semble the former immigrants. Our species, Homo sapiens, evolved

from its ancestors, Homo erectus, and perhaps other species as yet un-

discovered, in an extensive territory, comprising perhaps most of the

Old World. Evolutionary improvements, that is, new and adaptively

superior genotypes, arose from time to time in various parts of this

territory. The populations in which these improvements arose ex-

panded and transmitted their advantages to more widespread popula-

tions. Where two genetic improvements met, new populations of still

superior adaptedness were formed, and expanded in turn.

To ask where Homo sapiens first appeared is therefore meaningless.

Races and local populations are evolutionary trial parties which ex-

plore the various possibilities of adaptation. The gene pool of the

now living mankind contains genetic elements which were present in

many and perhaps in all major populations of the past.

Development of the Brain as the Moving Force of Human Evolu-

tion. However incomplete our knowledge of human ancestry, there is

scarcely any doubt that the development of brain power, of intelli-

gence, was the decisive force in the evolutionary process which cul-

minated in the appearance of the species to which we belong. Natural

selection has brought about the evolutionary trend towards increasing

brain power because brain power confers enormous adaptive advan-

tages on its possessors. It is obviously brain power, not body power,

which makes man by far the most successful biological species which

living matter has produced. The unprecedented and unparalleled

success of man as a species has led to an increase in the numbers of

living individuals from perhaps some hundreds of thousands during

the Ice Age to about two and a half billion at present. Man has

spread and occupied all continents and major islands except, perhaps,

the interior of Antarctica. He has destroyed or reduced to insignifi-

cance other organisms which were his competitors, or which preyed

on him as predators or parasites. He has domesticated many animal

and plant species, made them serve his needs, and changed them

genetically to improve their serviceability to him. Finally, man is in

the process of the acquisition of knowledge which may permit him

to control his own future evolution.

The human skeleton, and particularly the skull, underwent many

changes during the evolutionary transition from ape-like ancestors

(such as the Australopithecinae) to modern man. At first sight some

Development of the Brain 335

of these changes, such as those in the shape of the skull, the presence

of brow ridges, or in the presence or absence of a chin, may appear

fortuitous and without meaning. Actually, careful comparative anal-

ysis of these traits in different races and species shows that most of

them are part and parcel of the basic trend—the development of the

brain. To quote Weidenreich: "One of the most impressive experi-

ences a student of human evolution can have is to realize the extent

to which all the smaller structural alterations of the human skull are

correlated with and depend upon each other, and the extent to which

they are governed by the trend of the skull transformation as a

whole."

Examples shown in Figures 13.5 and 13.6 illustrate what is meant by

such correlations. The brain of an Irish wolfhound, a large dog breed,

is only about twice as heavy as that in the King Charles spaniel; the

weight of the body, however, is about ten times greater in the large

than in the small dog. In other words, a small dog has a relatively

much larger brain than a large one. The relatively large brain can

be accommodated in the small skull only by reducing the parts other

than the brain cavity. According to Weidenreich, transitions between

the gorilla skull, that of Java Man, and that of modern man entail

changes strikingly parallel to those found in large and small dogs.

Of course, man is not strikingly smaller than the gorilla in body size,

but his brain is both absolutely and relatively larger. A greater size

of the brain does not necessarily prove a higher intelligence. Among

living men the brain size varies greatly, and there is no strict relation

between its size and intellectual capacity. Among the great writers,

the brains of Jonathan Swift and of Ivan Turgenev measured about

2000 cubic centimeters each, while that of Anatole France was only

1100 cubic centimeters in volume. This does not contradict the fact

that groups of animals with larger brains show a higher intelligence

on the average.

The trend towards increasing brain size has played an important

role in the evolution not of man alone but of the whole primate order.

The sequence of forms beginning with the tree shrews and going to

other prosimians, monkeys, apes, and man, is characterized above all

by a growth of the brain and of intelligence. There have been other

evolutionary trends among the primates interacting with the brain de-

velopment. Washburn (1951) regards the achievement of erect pos-

ture and of efficient bipedal locomotion a critically important stage of

human evolution. To be able to walk and to run on the ground, in-

stead of climbing on tree branches, our ancestors had to modify their

•336

Human Evolution

w

Figure 13.5. Skulls of various primates. A, Notharctus, a fossil lemuroid; B,

Tetonius, a fossil tarsioid; C, Mesopithecus, a fossil Old World monkey; D, chim-

panzee; £, Australopithecus; F, Java man; G, Neanderthal man; H, Cro-Magnon

man. (From Colbert.)

338 Human Evolution

The Humanity of Man. It is hard to tell at what point in the evo-

lution of the hominoid stock the prehuman animal became a human

being. To some extent this is a matter of definition of what consti-

tutes a "human being." According to Le Gros Clark (1953) the

segregation of the family Hominidae (including Australopithecinae)

from the ape family (Pongidae) occurred in the Miocene, ten or more

million years ago. The most ancient and primitive tools (eoliths)

appear at the transition from the Pliocene to the Pleistocene periods

in Europe. However, the eoliths are so much like ordinary stones

that their nature is a matter of controversy. Unmistakable stone arti-

facts date from early to middle Pleistocene, perhaps close to a million

years ago, and more than 100,000 years ago expertly made tools were

available.

Man is not simply a very clever ape, but a possessor of mental abili-

ties which occur in other animals only in most rudimentary forms, if

at all. Many animals utter sounds as warning signals or as manifesta-

tion of emotions. A clog may bark, howl, growl, and whine in many

different ways, the meaning of which may be comprehensible to his

master and to other dogs and human beings. But only man uses words

to express concepts or to designate categories of objects or acts. A

word, after all, is a noise which the human larynx and mouth are able

to produce. But to become a word this noise must be invested with

a conventional meaning. Thus the sounds of the word "table" are in

no way descriptive of the object they denote; they become so by

virtue of a certain group of people having learned to associate them

with definite objects. Furthermore, the use of sounds for objects in-

volves abstract thinking: "table" means not only an individual piece

of furniture but many and diversified pieces having certain properties

in common.

Cultural Heredity. Man has developed cultural heredity or culture.

There are several definitions of culture. One of them is: "The total

life way of a people, the social legacy the individual acquires from

his group. Or culture can be regarded as that part of the environ-

ment that is the creation of man" (Kluckhohn). The essential thing

about culture is that it has to be acquired by each individual by

learning from others; it is not transmitted from parents to offspring

through the sex cells as is biological heredity. For example, nobody

is born able to speak, read, and write any language. Children have

to be taught speaking, reading, and writing.

And yet human infants are born able to perform the complex series

of muscular movements needed to obtain milk from their mothers'

Cultural Heredity 339

breasts. This is a biologically inherited behavior pattern, an instinct.

Many animals have wonderfully complex and efficient behavior pat-

terns of courtship, copulation, and care of offspring. These behavior

patterns need not be learned. They are set by the genes. Of course

cultural and biological heredity are not isolated from each other; they

constantly interact. For example, although we have to learn reading

and writing, the ability to learn is genotypic. The human sexual

drive, though instinctive at base, is overlaid with so great a mass of

culturally acquired conditionings and elaborations that it is chiefly

non-genetic in its manifestations.

Culture is acquired, and like all acquired traits it is not transmitted

through genes in the sex cells. This may at first glance seem to

diminish the adaptive effectiveness of culture. In reality the opposite

is true: culture is a powerful means for controlling human environ-

ments exactly because it is not biologically inherited. We inherit our

genes only from parents and other direct ancestors. We transmit

them only to children and to direct descendants. There is no way to

transmit our genes to even the most dearly beloved friends. The

biological heredity of populations can be changed only by the rela-

tively slow process of breeding and selection. By contrast, culture

can be transmitted to any number of contemporaneous individuals or

to future generations, regardless of biological descent or relationships.

The founders of great religions, scientists, inventors, poets, philoso-

phers, and men of action have influenced the cultural heredity of

mankind for many generations and perhaps forever. Cultural evolu-

tion is vastly more rapid and efficient than biological evolution (see

further discussion of this in Chapter 14).

Because of the great efficiency of the transmission of culture, man

has been able to master an immense variety of environments. Or-

ganisms other than man become adapted to their environments by

changing their bodies and their genes by the relatively inefficient

process of natural selection. Man alone adapts himself, in a large part,

by actively or even deliberately changing the environment, and by

inventing and creating new environments.

He uses his immense powers to modify the environment not always

wisely, but on a scale so grand that he has become an important not

only biological but also geological agent. He has plowed up the soil

and changed it by destruction of forests and other vegetation. The

adaptive advantage of the ability to acquire even the most rudimen-

tary forms of culture must have been so great in the early stages of

human evolution that natural selection rapidly propagated the geno-

340 Human Evolution

types which permitted the acquisition of culture throughout the human

species. The gene-controlled capacity to learn, absorb, and use new

techniques and tools was, then, developed, intensified, and diffused

by means of biological evolution, making our species more and more

human.

Rudiments of Cultural Transmission among Animals. Human intel-

lectual abilities seem to be not only quantitatively but also qualita-

tively different from those of animals other than men. It is important,

therefore, to demonstrate that rudiments of these specifically human

abilities can be found in certain animals. Natural selection found in

the ancestors of our species the raw materials from which the present

genetic endowment of mankind was compounded.

It can easily be observed that, at least among mammals and birds,

the offspring acquire certain behavior patterns by imitation and learn-

ing from their parents. Wolves and cats instruct their young in hunt-

ing techniques. But all this does not necessarily result in the forma-

tion of learned tradition comparable to human culture. In animals,

the individuals of one generation transmit to those of the next what

they themselves learned from their parents—not more and not less.

Every generation learns the same thing which its parents have learned.

In only very few instances the evidence is conclusive that the learned

behavior can be modified or added to, and that the modifications and

additions are transmitted to subsequent generations.

Promptov, Sick, and others found that some details of the song of

some birds vary in different local races, and that young birds learn

them by imitating their parents and other birds in the neighborhood.

(The ability to produce certain notes is, of course, a genetic trait of

each bird species, just as the ability to learn to read is a genetic trait

of the human species.) Three populations of the chaffinch, near

Stuttgart, Germany, each live in a definite neighborhood, and differ

in the calls which males utter in their breeding territories. One of

these populations is confined to a certain large park, which is known

to be about 300 years old. It is probable that the special "dialect"

of the chaffinches in this part has evolved within this period of time.

A spectacular instance of development of a "custom" has been ob-

served in the kea parrot (Nestor notabilis) in New Zealand. At some

time late in the last century the kea started to attack and kill sheep

and rapidly became a serious pest and an economic problem. Until

the arrival of the white man, New Zealand had no sheep and no

animals of similar size and character; the sheep killing is a new habit

which has spread among the kea parrots by imitation and learning.

Insect Societies 341

According to Lack, the finch Camarhynchus pallidus on the Galapagos

Islands uses a cactus spine which it carries in its beak to poke into

crevices of tree bark to get at insects on which it feeds. This animal

is, then, using a primitive "tool"; it would be interesting to ascertain

whether this tool-using habit is acquired by young birds by a process

of learning and whether it can be modified.

Brilliant experiments of von Frisch have shown that honey bees

have a "language" which, as recently pointed out by the anthropolo-

gist Kroeber, involves the use of some genuine symbols, which other-

wise are known for sure only in man. A bee which has located a

source of food returns to the hive and performs a special "dance,"

which imparts to the hive mates information concerning the direction

from the hive where the food is to be had. The movements of which

the "dance" is composed indicate the direction in a purely symbolic

fashion.

Insect Societies. Allce (1951) and Ashley Montagu (1950) main-

tain that all animals are social to some extent.. Indeed, even the re-

lationships between the sexes and between parents and offspring in-

volve some cooperation ("protocooperation" according to Allee) and

social behavior. Some animals, however, live in complex and highly

organized societies. Most monkeys and apes are social animals; they

live in bands or herds, warn each other of approaching danger, an-

nounce the finding of food, etc. It is probable that the development

of cooperation in the social life of prehuman primates was quite im-

portant in the origin of human mental abilities.

Apart from man, the most highly organized societies occur not

among mammals or birds but among insects—wasps, bees, and espe-

cially ants and termites. The study of the social life of these insects

has yielded some of the most fascinating stories which biology has to

offer. Ants and termites live in colonies which may consist of many

thousands of individuals. A colony builds a nest, often of elaborate

and species-specific architecture. Food is collected and stored in

special chambers. Larvae and different classes of adults may be pro-

vided with different diet. Some species of ants engage in "agriculture"

—they collect pieces of leaves, store them in their nests, seed them with

spores of special fungi, harvest the fungus growth when it reaches a

certain degree of maturity, and use it for food. Ants and termites

often have in their nests other kinds of insects or other animals which

stand to their hosts in a relation analogous to that of domesticated

animals to man. Some of these "guests" of the ant and termite nests

are unable to live independent lives, since they are fed and cared for

Innate and Learned Behavior 343

by their hosts. The benefits which the hosts derive from their charges

sometimes appear far-fetched; for example, some ant guests produce

odoriferous secretions which the hosts lick with avidity.

Most remarkable of all is the division of labor between the different

kinds, or "castes," of individuals in ant and termite colonies (Figure

13.7). A colony of ants has one or more "queens" and "kings," which

are females and males whose business is reproduction and who are

fed and cared for by the workers. Workers are sexually underdevel-

oped females who collect food and build the nest; soldiers, also under-

developed females, who take care of the defense from outside inva-

sions; and such curious castes as door-keepers with heads fashioned

into plugs to close the nest entrances, and living barrels, with enor-

mously swollen abdomens in which certain provisions are stored. All

members of a colony show complete "unselfishness" and "devotion"

to the common good.

Innate and Learned Behavior. In describing insect societies, it is

hard to avoid expressions like "unselfishness" and "devotion," bor-

rowed from the human social and ethical vocabulary. We must, how-

ever, beware of the fallacy of anthropomorphism, which ascribes to

animals human motivations and emotions. Yet some biologists have

again and again tendered the quaint advice, that human societies

ought to be reformed to emulate the "virtues" of insect societies. How

nice it would be if our society were organized for the benefit of all,

if every person was an expert in some useful function and performed

it to the best of his ability, and if all men were heroes ready to sacri-

fice themselves for mankind!

Such counsels are naive, because the behavior of social insects is

innate or instinctive, and is fixed by heredity; human social behavior

is learned from others or devised through reasoning and choice. It

is important to realize the implications of this distinction.

Innate behavior is a wonderful instrument which efficiently serves

the needs of the species in the environments in which this species

normally lives. The behavior of an insect building a nest or arranging

things for its progeny appears highly competent and wise. Most im-

portant, their behavior needs not to be learned. Among the marvels

of ant and termite societies one thing is conspicuously absent. No-

where is there a school for the young workers or soldiers! An ant just

Figure 13.7. Specialized individuals (castes) in ants. A, soldier; B, worker; C,

a winged male; D, a female who has lost her wings, of Pheidole instabilis; E, a

"replete" or honey barrel of Myrmecocystus hortideorum; F, a soldier door-keeper

of Colobopsis etiolata; and G, its head in front view. (After Wheeler, redrawn.)

344 Human Evolution

hatched from a pupa, or just having reached a certain stage of its

development, is just as expert at performing its work as it will be after

having done this work repeatedly. By contrast, human infants are

utterly helpless and must undergo prolonged training and education

before they can take care of themselves and become useful members

of society. In no other animal is this period during which the young

require protection and care so long as in man. But during this period

the young absorb the accumulated cultural experience of the human

species; they need not learn things only by trial and error or make

all the mistakes and discoveries which others have made before them.

However time-consuming may be the individual apprenticeship, a

very much longer process of cultural development is telescoped in

the upbringing and education (see Chapter 14).

There is an all-important limitation to the apparent wisdom of

innate behavior: it may no longer function to the advantage of either

the individual or the species in new environments, which the species

has seldom or never encountered in the past. The classical experi-

ments of Fabre (1823-1915) illustrate well the rigidity of instinctive

behavior. One example will suffice here. The mason-bee (Chalico-

doma) constructs pot-shaped cells from pellets of clay, fills them with

honey and pollen, deposits an egg on these provisions which the larva

hatching from the egg will eat, and finally seals the opening of the pot

with a clay cover. While the bee is in the process of constructing

the cell, she notices and repairs any damage which the construction

suffers. As soon as the building is finished and the insect begins to

collect the provisions, repairs are no longer made. Fabre made in

a cell a hole through which the honey then escaped. The bee went

on trying to fill the bottomless barrel. Similarly, while the provisions

are being collected, the insect removes and throws away any extraneous

particles that may happen to get into the cell. But when the egg is

laid and the opening is to be sealed, it is sealed regardless of the pres-

ence of conspicuous debris on the surface of the food.

This does not mean that instinctive behavior is absolutely fixed. In-

nate behavior patterns, like other traits determined by the genotype,

can be modified within certain limits by the environment. As always,

the genotype determines the norm of response to the environment.

Environmental variations which the species often meets evoke modi-

fications of the innate behavior patterns which are, as a rule, bene-

ficial to the organism. For example, the drive which makes birds of

a given species migrate in spring and in autumn to different countries

is innate, and is released by a physiological mechanism of a hormonal

Human Diversity 345

nature. But the birds do adjust their wanderings to some extent to

weather conditions. Similarly, ants can use a certain range of ma-

terials for the construction of their nests and adjust the shape of the

nest to local conditions. Innate behavior, however, is less plastic and

less versatile than learned behavior.

Human Diversity. Man inhabits all parts of the world and all

climes. He is consequently exposed to a great variety of environ-

ments. Moreover, he uses his ingenuity, his technological competence,

to invent ever new environments. Consequently, environments in

which men live change rapidly with time. To be sure, the diversity

and changeability of human environments are due chiefly to cultural

rather than to physical causes. The physical environments in which

men live are, in fact, becoming progressively more standardized. The

clothing which men wear, the houses in which they live, artificial heat-

ing, lighting, and assured food supply make the physical environments

perhaps more uniform now than they were during the formative period

of the biological evolution of our species.

Quite the reverse is true of social and cultural environments. Here

the diversity is enormous and ever increasing. Even in primitive

societies there is a division of labor among their members. Some indi-

viduals gather or produce food; others make implements and vessels;

still others act as priests, witch doctors, or leaders. In advanced

societies the functions to be performed are not only highly diversified,

but new ones constantly appear and the old ones vanish. Finally,

every individual performs different functions as a child, an adolescent,

an adult, and an old man or woman.

We know that a living species adapts itself to diverse and variable

environments by two methods (Chapters 6 and 7). First, a variety of

genotypes is produced, each specialized to fit a certain part of the

available range of environments. Second, genotypes are evolved which

permit their possessors to adjust themselves successfully to a certain

spectrum of environments by homeostatic modification of the pheno-

type. The first method involves genetic specialization; the second

emphasizes the adaptive plasticity of the phenotype.

The first method of adaptation makes the inhabitants of different

regions of the earth diverge genetically and form allopatric geographic

races. It also makes people living in the same region genetically di-

versified, and causes genetic variation between individuals and poly-

morphism in human populations. The second method makes people

able to adjust themselves to circumstances. Human plasticity and

educability make it possible to have most normal human beings trained

346 Human Evolution

to perform competently whatever functions the society may need to

have ministered to, although some persons may be genetically condi-

tioned to succeed in the performance of some functions more than of

others (see Chapter 14).

Classification of Human Races. The fact that human populations

of different countries are visibly and genetically different is, then, a

reflection of the diversity of environments in these countries. Human

races have developed through the adaptation of human populations

to their surroundings.

It has, however, been shown in Chapter 7 that the number of races

which one chooses to recognize in a species by giving them names is

largely a matter of convenience. It is not surprising, therefore, that

different authorities have held widely divergent opinions concerning

the number of human races.

Perhaps the simplest division of the human species is in three

major races: Negroid, Mongoloid, and White or Caucasoid. Negroids

have dark brown to black skins, frizzly or kinky hair, broad flat noses,

and usually thick lips. In Mongoloids the face is flattened, hair straight

and coarse, nose flat at the root and with moderately spread nostrils.

Whites have usually, though not always, pale skin, narrow and often

long noses, and straight, wavy, or curly hair. By distinguishing the

natives of Australia from the Negroids, the pre-Columbian inhabitants

of the Americas from the Mongoloids, and the inhabitants of the

South Sea Islands, from Hawaii to New Zealand, which are hard to

place in the three-race scheme, we arrive at the following six races:

(1) Negroid, (2) Mongoloid, (3) White or Caucasoid, (4) Australoid,

(5) American Indian, and (6) Polynesian.

The six-race scheme obviously fails to do justice to differences easily

observable among human populations. For example, the white race

of this scheme includes people as different as the predominantly blond

northern Europeans, the brunet Arabs, and the tawny Hindus. Very

naturally, anthropologists attempted to split the races further and

further in the vain hope of making them more uniform and more clear-

cut. The result has been a useless multiplication of racial names. The

dividing lines between the small races are often completely blurred

by the intermarriage and gene exchange which have been going on

for centuries and millennia and are becoming more and more frequent

as time goes on. Moreover, no classifications satisfactory to all anthro-

pologists has been arrived at; different authorities have proposed

different numbers of races.

A reasonable proposal, made in 1950 by three outstanding American

Classification of Human Races 347

anthropologists, Coon, Garn, and Birdsell, recognizes thirty races,

listed in Table 13.3 and Figure 13.8. Without attempting to describe

the races in any detail, some interesting features of this classification

deserve to be mentioned.

In the first place, Coon, Garn, and Birdsell claim no finality for their

thirty-race classification. They recognize that race is not a static but

a dynamic entity. Old races may disappear, either because human

populations may die out or because races may merge together owing

to frequent intermarriage. New races are formed by selection, hybridi-

zation, and genetic drift. Some of the races listed in Table 13.3 did

not exist a thousand years ago; others were relatively more numerous

than now; many have become distributed more widely, while some

have contracted the areas of their habitation. Thus the Murrayian

race in Australia (1) and the Ainu race (2) in eastern Asia are on

the verge of extinction or of being engulfed by intermarriage with

other races. Both these races showed high frequencies of heavy brow

ridges, large teeth, and thick skull bones, which are traits often met

with in early human races and species known as fossils (see above).

Lapps (6), Negritos (9), Bushmen (10), Carpentarians (13), and

Dravidians (14) may also be engulfed by the more numerous races

near whom they live.

On the other hand, the beginnings of the formation of the races

(19), (20), (28), and (30), Table 13.3, date back only some centuries.

The North American Colored race (19) is a result of admixture of

genes of European origin, and of some American Indian genes, in the

gene pool formed by a mixture of several races native to Africa. An

analogous population is South African Colored (20) which arose by

hybridization of races (10) and (11) with immigrants from western

Europe. The Ladinos (28) are a series of populations which exist

as castes or classes in some Latin American countries; they arose by

hybridization of the immigrant Mediterraneans (17) with the native

Indians (27, some 26), and, in places, some Negroes. Finally, the

Neo-Hawaiians (30) are a very recent population arising from hy-

bridization of the Polynesian race (29) with migrants from various

parts of Asia, America, and Europe.

These new races are biologically no less "real" than the old ones

which have contributed to their formation. No race of a sexual and

cross-fertilizing species ever consists of genetically identical indi-

viduals. It cannot be too often emphasized that "pure races" exist

only in asexual organisms and are figments of the imagination as far

as man is concerned. Gene exchange between races of a sexual spe-

Classification of Human Races

349

TABLE 13.3

RACIAL CLASSIFICATION OF MANKIND, ACCORDING TO

COON, GARN, AND BIRDSELL

Name of the Race

1. Murray ian

2. Ainu

3. Alpine

4. Northwest European

5. Northeast European

6. Lapp

7. Forest Negro

8. Melanesian

9. Negrito

10. Bushmen

11. Bantu

12. Sudanese

13. Carpentarian

14. Dravidian

15. Hamite

16. Hindu

17. Mediterranean

18. Nordic

19. North American

Colored

20. South African

Colored

21. Classic Mongoloid

22. North Chinese

23. Southeast Asiatic

24. Tibeto-Indonesian

Mongoloid

25. Turkic

26. American Indian,

Marginal

27. American Indian,

Central

28. Ladino

2!). Polynesian

30. Neo-Hawaiian

Country of Origin or Residence

Aboriginal population of southeastern Australia

Aboriginal population of Japan

Central Europe to western Asia

Native in northern and western Europe, now world-

wide

Russia and Poland

Northern Scandinavia

Western Africa and Congo

New Guinea to Fiji and New Caledonia

Enclaves in Africa, Philippines, New Guinea, Anda-

man Islands

Aboriginal population of South Africa

East and South Africa

Upper Nile, Sudan

Aborigines of northern and central Australia

Aboriginal population of southern India

East Africa and Sudan

India

Southern Europe, North Africa, Near East

Northern and western Europe

The "colored" population of United States

The "colored" population of^outh Africa

Eastern Siberia, Mongolia, Korea, Japan, Eskimo

Northern and central China

Southern China, Siam, Indonesia, Philippines

Tibet, northern Burma, parts of Indonesia

Turkestan, Central Asia

Most of American Indians

354 Human Evolution

is known much better, and this allows some insight to be gained into

the processes of race formation and amalgamation.

Adaptive Nature of Human Races. Man's most powerful means of

adaptation to his environments is learned and acquired knowledge,

the ability to choose consciously and to create new environments, in

short his culture. In the concluding chapter of this book it will be

argued that the genetic endowments which permit the acquisition and

maintenance of culture are the property of the human species as a

whole, not of any one race. Relative to the efficacy of cultural adapta-

tion, the importance of the genetic differences between the races is

very small indeed. It is also diminishing with time, as civilization

discovers ever new methods of controlling environments. In fact,

surprisingly little is known about the adaptive significance of the

particular traits which distinguish human races. Coon, Garn, and

Birdsell (1950) and Coon (1954), however, have made some tentative

suggestions which may be reviewed here.

People with black or chocolate-brown skin are native to the forests

and grasslands of Central Africa, Melanesia (New Guinea and islands

to the southeast), southern India, and some islands on the fringe of

southern Asia. All these regions are located in the torrid zone along

the Equator, where the sunlight is intense and the danger of sunburn

is great. The skin pigment absorbs the ultraviolet radiation which is

responsible for the sunburn. No dark-skinned population has yet de-

veloped in the American tropics, partly because human occupancy of

these lands is relatively recent, and partly, according to Coon, because

most populations there live either in the shade of dense forests or in

the mountains. People native in temperate and cold climates can

develop the protective tan following skin exposure to light, or can

bleach if covered with clothing or if the weather is cloudy. Bleached

skin is supposedly advantageous because it permits the scarce ultra-

violet rays to produce enough of the vitamin D, the "sunshine vitamin"

which is necessary for health. Skin which can either darken or bleach

is, therefore, advantageous where the amount of sunshine is variable

with the season (cf., however, Chapter 7).

In bodies of similar shape, the surface grows as the square, and the

volume (or weight) as the cube, of the linear dimensions. In other

words, the smaller the body, the greater its surface in relation to the

volume. Body surface is relatively greater in tall and slender people

than in short and rotund people of similar bulk; long arms and legs

and large flat hands and feet also increase the body surface. The body

surface acts as a radiator of an automobile does, causing dissipation

Suggestions for Further Reading 355

of heat and cooling of the body. It follows that an increase of the

body surface relative to the body mass is desirable in hot climates,

and a decrease in cold climates. There is some evidence that the

observed racial variation is on the whole in accord with these require-

ments. In Europe the bulkiest people inhabit the North, whereas the

Mediterranean people tend to be more slight and gracile. Among

the Mongoloids, northern Chinese are larger people than their south-

ern countrymen, Annamites, and Siamese. In the Americas, the Maya

of southern Mexico and Guatemala, and the Amazonian Indians are

small, whereas the Indians of Alaska, Canada, the northern United

States, and of southern Argentina are more bulky. In Africa, the

Nilotic Negroes are among the tallest people in the world, but they

are also exceptionally slim. Coon is also of the opinion that the char-

acteristic features of the Mongoloid face may represent an adaptation

to the cold, dry, and windy climates of the great interior of Asia.

Not all racial traits need be so utilitarian as suggested above. Some

facial features, hair shapes, and perhaps body forms may have become

established by natural selection in response to the vagaries of popular

tastes and ideals of bodily beauty. The possessors of traits considered

comely and pleasing may have been favored as mates and placed in

superior positions to raise large families. Much study and research

are necessary before the origin of human racial characters can be

fully understood.

Suggestions for Further Reading

Darwin, Ch. 1871. The Descent of Man.

This book remains a great classic of evolutionary literature.

For more up-to-date information on human origins consult the following five

books:

Coon, C. S. 1954. The Story of Man. Knopf, New York.

Howells, W. W. 1945. Mankind so Far. Doubleday, New York.

Howells, W. W. 1954. Back of History. The Story of Our Own Origins.

Doubleday, New York.

Ashley Montagu, M. F. 1951. An Introduction to Physical Anthropology. 2nd

Edition. Ch. Thomas, Springfield, 111.

Weidenreich, F. 1946. Apes, Giants, and Man. University of Chicago Press,

Chicago.

Concerning human origins, as well as concerning the problems of human race

and racial classification, consult also the books of Count, Boyd, Coon, Garn and

Birdsell, and the Symposium on the Origin and Evolution of Man, cited among

the suggested readings in Chapter 7.

356 Human Evolution

The story of social insects and the relations between the innate (instinctive)

and the learned behavior are discussed in the following books:

Allce, W. C., Emerson, A. E., Park, O., Park, Th., and Schmidt, K. P. 1949.

Principles of Animal Ecology. Saunders, Philadelphia.

See particularly Chapters 23, 24, and 35.

Haskins, C. P. 1945. Of Ants and Men. Allen & Unwin, London.

Haskins, C. P. 1951. Of Societies and Men. Norton, New York.

Lorenz, K. 1952. Ktng Solomon's Ring. New Light on Animal Ways. Crowell,

New York.

Chance, Guidance, and

Freedom in Evolution

The idea of evolution, of transformation of one kind of organism

into another, certainly antedates Darwin. Before scientific biology

appeared, even the weirdest stories of transformation were often

credited (see page 166). In classical antiquity, the creation myths

of Anaximander, Empedocles, and Lucretius fancied that living beings

arose from very different progenitors and from inanimate matter

(page 111). H. F. Osborn (1857-1935) and others claimed that

Aristotle was also an evolutionist, since he maintained that nature

advances from the inanimate to the animate, and from less perfect to

more perfect creatures (page 224). It is not, however, certain whether

Aristotle meant this advancement as a concrete historical event, or only

as a part of his more general philosophical view that a vital force, a

soul, gives a recognizable actuality and "form" to potentiality and

"primary matter." Aristotelian views were taken over by medieval

philosophers and theologians, particularly by St. Thomas Aquinas,

who did not interpret them in any recognizably evolutionist sense.

On the other hand, Descartes (1596-1650) and Buffon (1707-1788)

apparently did arrive at evolutionist views of nature. But in their

day such views were regarded as subversive, and Descartes and Buffon

accepted the dictates of authority and were in no mood to risk their

privileged positions. They saw fit to disguise their views as mere

amusing paradoxes or idle play of the intellect. Maupertuis (1698-

1759), Erasmus Darwin (grandfather of Charles, 1731-1802), Goethe,

and several others mentioned in Chapter 10, approached the problem

of evolution in various ways. The possibility that the world of life

might be a product of evolution was, indeed, "in the air" when Charles

Darwin started his work.

357

358 Chance, Guidance, and Freedom in Evolution

The greatness of Lamarck and of Darwin was that they not only

adduced new and compelling evidence of evolution (which they cer-

tainly did), but that they also made the occurrence of evolution in-

telligible. Lamarck and Darwin agreed that the evolution of life on

earth had been brought about by causes which continue to be in opera-

tion even now. These causes can, then, be observed and experimented

with. It happens that a century and a half of observations and ex-

periments have failed to confirm Lamarck's surmise of what these

causes are. Darwin's ideas have stood the test of time much better.

Both Lamarck and Darwin were innovators who proved that life has

had history, and that this history is not unfathomable. Although we

shall never be able to have the evolution of the living world re-enacted

as a whole before our eyes, at least some elementary evolutionary

events have been reproduced (Chapters 5 and 9). Man may yet learn

to direct the future course of the evolution of species, including his

own.

The general biological theory of evolution outlined in this book

developed, by an unbroken continuity of thought, from the Darwinian

prototype (Chapter 6). However, biology has not stood still since

Darwin; therefore, the modern theory differs greatly from Darwin's.

Not all biologists, however, are satisfied that this theory is valid.

Quite properly, they have tried to suggest alternative possibilities,

and these alternatives should be given consideration. Even though

none of them may prove acceptable, a clearer perspective of evolution-

ary biology will be gained in the process.

Autogenesis and Finalism. The longing to know the future is

deeply ingrained in human nature. Science helps to understand the

present, to comprehend the past, and to predict the future. Astrono-

mers can foresee the second when a sun eclipse will be visible from

any given point of the earth's surface, and can describe eclipses which

happened centuries ago. The astronomer Laplace (1749-1827) as-

serted that, given for one instant a knowledge of the positions and the

velocities of all the masses composing the universe at any particular

moment, "an intelligence vast enough to submit these data to analysis

would embrace in the same formula the motions of the greatest bodies

in the Universe and those of the lightest atoms; for such an intelli-

gence nothing would be uncertain, and the future, as well as the past,

would be present to its eyes." Of course, no human intelligence is

anywhere near so vast, but Laplace's proud hope became a basic tenet

of mechanistic science.

The present state of the universe, then, is the consequence of its

Autogenesis and Finalism 359

previous states and the cause of its future states. Once this is granted,

it inevitably follows that the evolution of life, of man, and indeed of

the whole cosmos, was predetermined and fixed from the start. This

conclusion harmonizes very nicely with the biological preformation

theory (Chapter 10). Primordial life contained within itself all the

evolution to come, as a flower bud contains all parts of the future

flower. Evolutionary changes arise, some biologists suppose, from

inside the organism, autogenetically, and lead only to the unfolding

of what was there from the beginning. It was not natural selection

at all which has shaped the diversity of living creatures. If evolution

is brought about by autogenesis, then natural selection can accept

only what is fit to survive in a given environment and reject what is

unfit. An analogy will make clearer what is meant by autogenesis.

The element uranium decomposes into helium and lead; the rate of

decomposition is constant and largely independent of the environment;

it is possible to predict that after a certain number of years a certain

fraction of the uranium will have turned to lead. May organic evo-

lution be caused by a similar process? L. S. Berg (1926) thought that

it might. According to him, evolution is "nomogenesis," that is, "devel-

opment according to law" residing in the living matter itself. What

evolution produces is what it is destined to produce, just as a flower

bud is destined to produce a flower. Rosa's "hologenesis" (1931) is a

similar notion.

We need not be mechanistic materialists to believe in autogenesis.

A modern brand of vitalism, known as finalism, credits this nolion.

Cuenot (1941) and Vandel (1949) in France, their popularizer Du

Notiy (1947) in the United States, Naef (1919) and Dacque (1931,

1940) in Germany, and other finalists have no use for crude material

forces such as mutation and selection. The problem of evolution is

resolved more simply. Evolution strives to reach a predetermined

goal or end. This goal is assumed to be the production of man. To

Dacque, the amoebae, fishes, amphibia, etc., are "disguised states of

mankind," whatever this may mean. To Vandel, "Matter was at its

origin rich in power and in unrealized possibilities, able to engender

the organic as well as the inorganic, the living and the inert. In

giving birth to life, matter passed to the latter the essence of its creative

energy, and this effort reduced matter to a degraded relic, devitalized,

and having lost most of its ancestral qualities." Having produced man,

the evolution of life has done its job, and has terminated. Organisms

other than man are "by-products, slags, left over after its (man's)

production." H. F. Osborn (1934) held opinions close to those of the

360 Chance, Guidance, and Freedom in Evolution

modern finalists. To him, the important principle which brings forth

evolution was "aristogenesis," an urge towards greater "perfection,"

assumed to be inherent in life.

Orthogenesis. To most biologists the speculations of the finalists

and other believers in autogenesis seem frankly puzzling. Neverthe-

less, the validity of a scientific theory, however startling it may seem,

is tested by its ability to account for facts which are incompatible with

other theories. The alleged basis of autogenetic theories is the appar-

ent directedness of evolutionary changes, revealed especially by the

fossil evidence. Consider the succession of the shell forms in the

snail shown in Figure 12.1. In each successive geological stratum the

shells become more and more angular, as though the evolution of these

snails were directed by some force straight towards attainment of

the greatest possible angularity. For such apparently directed trends

of evolutionary changes Eimer (1897) introduced the word "ortho-

genesis."

One of the most famous examples of orthogenesis is the develop-

ment of gigantic antlers in the males of the Irish elk, Megaloceros

(Figure 14.1). Starting with late Tertiary times (Pliocene) the an-

cestors of Megaloceros were getting progressively larger in size, and,

as they were getting larger, the antlers in the males were getting more

and more enormous, until during the Ice Age the antlers reached seem-

ingly absurd dimensions; and finally these animals died out. Some

paleontologists surmised that such huge antlers must have been in-

jurious to their carriers, and concluded that Megaloceros died out

because its antlers got too big for it to carry. In other words, the

evolutionary trend towards large antlers developed such a momentum

that it could not stop, even when it got to be harmful and led to

extinction. Simpson's fitting comment about this surmise is: "I can

only feel awe for any one who knows that structures were disadvan-

tageous in animals that were very abundant for tens of thousands of

years and more."

The evolutionary history of the horse tribe, considered briefly in

Chapter 12, has often been alleged to represent another clear instance

of orthogenesis, and has been cited as such in many books on biology

and evolution. Indeed, starting with the little Eohippus, the horses

were getting larger and larger, the modern thoroughbred horse (Chap-

ter 9) being a veritable giant compared to Eohippus. Furthermore,

beginning with the Condylarths and the Eohippus, the ancestors of

the horse were losing toe after toe, until the modern horses have just

a single toe left on each foot to stand on. The teeth became pro-

362 Chance, Guidance, and Freedom in Evolution

sided over by its function. This is an orthogenesis of sorts. But does

it follow that the Australopithecinae, or the tarsiers, or any other

human ancestors or relatives were men in disguise, who needed mil-

lions of years to slough off their animal masks? There is nothing to

necessitate such a view. Not all primates, nor all anthropoids, evolved

in the direction of man, and there is no reason to think that they will

do so in the future.

Orthoselection. The above examples of orthogenetic trends are

not exceptional. Fossil histories of many groups of organisms may be

interpreted as showing trends of one kind or another. Many paleon-

tologists and some biologists believed that a theory of evolution of the

type outlined in the foregoing chapters of this book could not explain

the apparent prevalence of orthogenetic trends. If so, we would have

to turn to one of the theories with names ending in—genesis mentioned

above. The German paleontologist Schindewolf (main works in 1936

and 1950) has been particularly emphatic in expounding the view

that an explanation on autogenetic or finalistic lines is necessary.

Simpson (1944, 1953) and Rensch (1947) have performed a great

service to science by demonstrating that there is actually nothing in

the known fossil record to contradict the modern biological theory of

evolution.

In speculating about the evolutionary trends disclosed by the fossils,

we must be on guard in order that, seeing the obvious, we do not miss

the significant. Simpson, having analyzed the fossil evidence of

horse evolution in greater detail than any one else, has concluded that

the orthogenetic interpretation can here be sustained only by disre-

garding important facts. As shown in Chapter 12, there were not

one but many evolutionary lines of horses, and not all of them by

any means were getting bigger all the time, or getting rid of toes, or

developing teeth fit for grazing. True enough, these things happened

in some lines of descent, including the one which gave rise to the

horses which are now alive. The horses which were largest and strong-

est, which could run fastest, and which were able to utilize the most

abundant food supplies by grazing on grass, survived, while those

animals which did not develop these characteristics died out. On the

whole, the fossil evidence lends no support to the idea that there was,

or is, a built-in propensity to develop in any particular evolutionary

direction. It is quite consistent with the view that the evolutionary

changes took place owing to natural selection of genotypes which

were most suited to exploit certain environmental opportunities, par-

Orthoselection 363

ticularly living on open grassy plains which were becoming widespread

on earth during the Tertiary geological period.

Of course the evolutionary changes did go in the same directions and

often for prolonged periods of time in many groups of organisms.

Moreover, the same or parallel evolutionary trends are often ob-

served in different organisms. Perhaps the most widespread and best-

known trend concerns gradual but steady increase in body size. This

growth has occurred in quite different animals, making them grow

larger and larger as time went on. The most ancient mammals (living

during the Mesozoic times, see Chapter 12) were only as big as mod-

ern mice or, at most, rats. Larger quadrupeds appeared during the

Tertiary. The known history of the elephants begins in late Eocene

times with animals about as big as a hog. As early as the Oligocene,

forms appeared that were as large as a bull. During Miocene and

Pliocene came elephants as large or larger than the modern ones,

and the giants like the woolly mammoths lived during the Ice Age

and died out, perhaps with the active assistance of man who must

have regarded them as excellent sources of meat. The increase of

the body size in horses has already been discussed, as well as the body

size progression among the primates, which may have been our own

ancestors (Chapter 13). The primitive reptiles of the Pennsylvanian

and Permian ages were only as big as some big lizards, yet they appear

to be the ancestors of the most gigantic land animals ever produced,

which lived during the Jurassic and Cretaceous periods (see Chapter

12). Evolutionary size increases have been observed also in some

invertebrate animals, particularly in mollusks.

There is, however, no iron-clad law that would make evolution

produce ever larger animals. For one thing, many animals have re-

mained as small as their ancestors, and some even became smaller.

For another, the prevalence of the tendency towards bigness is quite

understandable, since larger individuals are usually also stronger, bet-

ter able to resist some enemies and escape from others. In addition,

a more massive body is advantageous in cold climates (see page 355).

No wonder, then, that natural selection by and large favors bigger

animals over smaller ones. The selection becomes, according to

Simpson, orthoselection, which operates generally in the same direc-

tion. Orthoselection brings about the appearance of orthogenesis.

The basic difference is here that Orthoselection lasts only as long as

the environmental opportunity which favors it; orthogenesis has usu-

ally meant a change independent of, or even contrary to, such oppor-

tunity. Orthoselection is simply selection long continued.

364 Chance, Guidance, and Freedom in Evolution

Allometric Growth. A human infant has a relatively much bigger

head and smaller legs and arms than an adult person. This amounts

to saying that as our bodies grow, the different organs and parts do not

grow in the same proportion. The head grows less than the extremi-

ties, so that an adult is not just like a baby magnified, nor is a baby

a diminutive adult. Such difference in relative growth rates of body

parts is known as allometry or allometric growth. Allometry occurs

in many kinds of animals and also in plants. Allometry can be ob-

served also when adult bodies of different species or races are com-

pared. Larger species and races often show predictably different

body proportions, as illustrated in Figure 13.6 for the skulls of the

larger and smaller breeds of dogs.

Allometry is a reasonable explanation of some otherwise very puz-

zling evolutionary changes. Consider again the Irish stag Megaloceros

(Figure 14.1). This extinct form had not only much larger antlers

than its ancestors and relatives, but it was a larger animal. The evo-

lution of the ancestors of Megaloceros brought about an increase in

body size as well as in antler size. Now among deer and their rela-

tives, as the body grows larger, the antlers increase in size even faster.

If natural selection for any reason favored larger and more powerful

bodies, the antlers would be expected to overtake the body in the

rates of increase. This prize example of orthogenesis (see above) is

most likely an instance of orthoselection. Seeing the monstrous antlers

of Megaloceros we wonder why selection should ever have created

such an unwieldy structure. But it is as certain as such things can be

that natural selection does not bring into being antlers separate from

heads, or heads separate from the shoulders, etc. It is an individual's

body as a whole that lives, grows, reproduces, and dies. And natural

selection is opportunistic: So long as bigger stags transmitted their

genes to the following generations more efficiently than smaller stags,

the selection favored increasing size. Assuming that the huge antlers

were a burden to the animals (and this is an assumption which may

easily be wrong), a genotype which would have produced a bigger

body with relatively smaller antlers would have had an advantage.

There is, however, no warrant to believe that any combination of

relative sizes of body parts will always be ready when circumstances

favor it. This belief would imply a miraculous prescience of the

future on the part of the genotype (see page 107), and genes are, after

all, things of this imperfect world.

Evolutionary Youth and Senescence. The way of all flesh is from

birth, childhood, exuberance of youth, vigor of maturity, senility to

Evolutionary Youth and Senescence 365

inevitable death. The fossil record shows group after group of or-

ganisms which appear, become diversified and abundant, and then

decline to extinction. An analogy is tempting: Are youth, senescence,

and death the destiny of species, genera, and orders, as they are of

individuals? Some believers in autogenesis-orthogenesis thought so.

But analogies are a precarious method of scientific cognition. Growth

and senescence of individuals are brought about by physiological

causes and presumably do not involve changes in the genes; evolution

is genetic change. The two things cannot be equated hastily.

Let us see what is meant by evolutionary youth and senility of a

group of organisms. The reptiles were "born" in Pennsylvanian times

as an offshoot from the amphibians, and the "infant reptile," called

Seymouria, was so amphibian-like that, according to Romer, "the aca-

demic question has been raised as to whether Seymouria was an

amphibian which was almost a reptile or, on the other hand, a reptile

which had just ceased to be an amphibian." Very soon (geologically

speaking), namely during the Permian, there were present representa-

tives of several (six) orders into which the class of reptiles is divided.

Several more orders appeared during the Triassic, including the small

ancestors of the future giant "ruling" reptiles. This, then, is the

"youth" of the reptiles. The Age of Reptiles, when they were most

numerous and diversified, and when the gigantic forms were living,

occurred during the late Triassic, Jurassic, and Cretaceous. This is

"maturity." But, remarkably enough, no new orders of reptiles ap-

peared when as a class they were dominant. The "senility" came with

a crash in the late Cretaceous and in early Tertiary, when all but four

previously inconspicuous orders (turtles, crocodiles, lizards and snakes,

and the order of the New Zealand tuatara, or Sphenodon) died out.

These four orders are still hanging on, but the class of reptiles as a

whole is "senile."

Mammals have a similar story. They appeared as an offshoot of the

reptiles when the latter were "young." The "infancy" of the mammals

was a prolonged one, extending through the whole Age of Reptiles;

their vigorous "youth" came in the early Tertiary (Paleocene and

Eocene), coincidentally with or immediately after the dying-out of

the ruling reptiles. Most living orders of mammals, and some of the

extinct ones, appeared at that time, inaugurating the Age of Mam-

mals (see Chapter 12). The greatest blooming of the class of mam-

mals, when they were most diverse and included some spectacularly

large or peculiar forms, came in late Tertiary (Miocene and Pliocene).

A "senility" may have begun during the Ice Age and the Recent, when

366 Chance, Guidance, and Freedom in Evolution

there appeared a curious mammal called man. This upstart has suc-

ceeded, however, in making the present time the Age of Man.

The careers of the classes of reptiles and of mammals have nothing

in them that would compel us to assume that evolutionary "youth"

comes inescapably after "infancy," or "senility" after "maturity." De-

spite the fossil record's being incomplete and in many ways baffling,

we may entertain a working hypothesis that the evolutionary events

for which this record stands were brought about by natural selection

in response to environmental opportunity. Mammals lingered incon-

spicuously for many millions of years while the reptiles were dominant,

but underwent an exuberant adaptive radiation as soon as the domi-

nant reptiles died out. Was this just a coincidence? Is it not more

likely that the decline of reptiles opened up biological opportunities

which were quickly seized upon by mammals (and by birds) ? Again,

while most orders of reptiles died out, that of the lizards and snakes

suffered no eclipse and is doing nicely even now. They have a place

in the sun which is not much encroached upon by the mammals.

The birth of a new group of organisms usually means that an evo-

lutionary "invention" has been made which permits life to exploit

novel environments, or to exploit old environments in novel ways.

Thus amphibians were the first vertebrates to live on land; reptiles

became free from dependence on water as an abode of the early

developmental stages (tadpoles); mammals and birds evolved regu-

lation of the body temperature that makes them less dependent than

the reptiles are on the vfegaries of weather; besides, mammals "in-

vented" superior ways of taking care of their young, while birds be-

came able to exploit the food resources of the air (Chapter 12).

Clearly, natural selection would be expected to perpetuate such useful

"inventions." It would also be expected to encourage the adaptations

to the various aspects of environment, hence the adaptive radiation

of the evolutionary "youth" period (Figure 6.1).

Causes of Extinction. Extinction is a frequent finale to many evo-

lutionary histories. The causes which brought about the extinction

of many groups of organisms are quite obscure, which is not really

surprising, since, even if the fossil record were much more complete

than it is, we would often be unable to visualize all the manifold and

complex interrelations of the inhabitants of the remote past. But

the supposition that extinction occurs "obeying certain internal im-

pulses concealed in the constitution of the organism" (Berg) is barren

as a working hypothesis. A diligent study of the interrelations of now-

living organisms, however, throws some light on causes of extinction.

Causes of Extinction 367

Fossil remains of reptiles very much like Sphenodon occur in Tri-

assic and Jurassic rocks in different parts of the world. Then they

disappear from the record for at least 130 million years (Cretaceous

to Recent, Table 12.1). A population of Sphenodon is nevertheless

living on an isle off New Zealand, quite oblivious that it should have

long ago felt "internal impulses" towards extinction. However, it is

hardly an accident that the living Sphenodon is preserved in a locality

so far out of the way. The animal life of New Zealand is limited in

variety and lacks representatives of many kinds of creatures which

are common elsewhere. The competition of these creatures would

probably make short shrift of the Sphenodon. On a larger scale, the

marsupials are now restricted chiefly to Australia, where the placental

mammals were few until man came (Chapter 12). The marsupials,

with the notable exception of the opossum, have died out outside

Australia, presumably because they were no match for the placentals.

Every species is enabled by its body structure to occupy and exploit

certain ecological, or adaptive, niches in the economy of nature (page

311). Extinction occurs either because the ecological niche disappears,

or because it is wrested away by competitors. The ecological niche of

a parasite is its host; if the host species dies out, so does the parasite,

provided that it cannot victimize alternative hosts. The more narrowly

specialized is an organism for life in only certain environments, the

greater the risk of extinction. Evolutionary "senility" may be a con-

sequence of specialized adaptation to only a restricted ecological

niche. A narrow specialist may live quite happily so long as his

exclusive abilities find an assured outlet. The Teddy-bear-like mar-

supial koala (Phascolarctos, Figure 12.8) deigns to eat nothing but

the young foliage of only a few species of eucalypts. Nevertheless this

slow and defenseless animal was quite common in parts of Australia

until white men started to use it for target practice. Compare this

with the dietary versatility, aggressiveness, and watchfulness of such

an animal as the gray rat. Koala is now kept in existence only by

conservationist laws, whereas the rat is far from rare despite deter-

mined campaigns of extermination.

It may seem surprising that evolution controlled by natural selec-

tion leads so often to overspecialization and consequent extinction.

But this is only a consequence of the fact repeatedly emphasized above

that natural selection is opportunistic and, like any natural process

other than the human mind, lacks foresight. Selection perpetuates

what is advantageous here and now, and fails to perpetuate what may

be beneficial in the future unless it is also immediately useful. To

368 Chance, Guidance, and Freedom in Evolution

take up again the case of the unwieldy antlers of the stag Megaloceros.

Provided that larger stags produced more surviving progeny than

smaller ones, there must have been an "orthoselection" for size in-

crease. It is, then, conceivable that because of some environmental

changes the cumbersome antlers may have become injurious; if the

reversal of the size trend could not be accomplished quickly, the ani-

mal died out.

Origin of Complex Adaptations. A difficulty, fully realized by

Darwin, which any rational theory of adaptation has to face is the

formation of complex organs and physiological functions. Consider

the human eye with its many wonderfully coordinated parts, each part

necessary for the organ to be fully serviceable. Or consider the proc-

esses of pregnancy and childbirth. Here is a series of hormones, each

with a separate function, and yet acting in an ordered sequence like

different instruments in a symphony orchestra; even a slight disturb-

ance may make the process of reproduction end in failure.

Can such delicately engineered systems possibly arise through nat-

ural selection acting on the genetic variability supplied by mutation?

A mutation is, essentially, a "mistake" in the process of gene repro-

duction, and most such "mistakes" are detrimental to the organism

(Chapter 4). Yet these "mistakes" must be woven by natural selec-

tion into such patterns as give rise to useful organs and functions.

Let us keep in mind that selection has no foresight and cannot build

organs useful only in the future. An organ must be continuously use-

ful, or else selection will neither advance its construction nor even

maintain it. To some biologists this requirement seemed too great.

"We might just as well expect that if the wheels, screws and other

component parts of the mechanism of a watch were to be put into

a vessel, we could, by a simple process of shaking, get them to com-

bine in such a manner as to become a watch that would function as

such" (Berg). Theories of autogenesis have been urged on this basis.

The objection that theories of evolution by natural selection rely

too much on "blind chance" has been made repeatedly since Darwin's

day. Mutations are accidents of gene reproduction; gene recombina-

tion yields by chance various genotypes; selection improves the chance

that some genotypes will leave more offspring than carriers of other

genotypes. Can these chances add up to building a complex organ

such as an eye? The "watch analogy," however, contains a subtle

fallacy. It tacitly assumes that the human eye arose in all its present

perfection all at once, by a lucky throw of genetic dice. Such a sup-

position would, indeed, stretch our credibility to the breaking point.

Preadaptation 369

What actually happened was that eyes were gradually formed and

perfected among man's close and remote ancestors. These ancestors

had eyes for at least 400 million years (the appearance of the verte-

brates in the fossil record). Reptiles have eyes, and so do amphibians

and fishes, and even the lowly lancelet (Amphioxus) has pigment

cells which make it able to perceive light. Many of man's ancestors

had eyes which were simpler in structure and less perfect in function

than our own, yet the eyes were always useful to their possessors.

Natural selection had ample opportunity to press onward toward high

degrees of perfection of the eyes.

The most misleading implication of the "watch analogy" and of

similar arguments is that the eye was formed by summation of inde-

pendent mutations, each responsible for a certain part of the organ—

the lens, or the iris, or the rods or the cones of the retina, or the

muscles which move the eye, etc. If this were the true nature of

genes and mutations, the eye could not function until the last part of

the eye mechanism had been installed in its proper place. An auto-

mobile motor does not work until every component part is put where

it belongs. But genes and mutations do not work that way, and this

makes the difference. Genes determine not body parts but develop-

mental processes. The genes make a fertilized egg develop by stages

in a body which has, among other parts, eyes of a certain kind. Evo-

lution of the eye, then, is not like the building of a unit motor; it is

rather more comparable to the gradual development of internal com-

bustion engines from the first hesitant model to the present powerful

and efficient makes.

Preadaptation. Another dialectical tangle in which some evolution-

ists have ensnared themselves concerns the origin of genetic changes

which prove to be adaptive. As we have seen, mutations may be

described metaphorically as "mistakes" in the processes of the self-

synthesis of genes. These "mistakes" occur regardless of the need of

the organism to maintain or to improve its adaptedness to the environ-

ment. Nevertheless, some of them occasionally prove to be adaptive.

Such occasional mutations may be said to be "preadapted" to certain

environments actually before the organism has had any chance to

make use of them. From this is only a short dialectical step to say-

ing, as some finalists and vitalists actually did say, that evolutionary

changes are in general preadapted to certain environments, whether

these environments do or do not actually exist in reality. Thus the

ancestors of the ancient amphibians were becoming "preadapted" to

exploit the food resources of land before they actually started to walk

370 Chance, Guidance, and Freedom in Evolution

on it (cf. Chapter 12, page 290); some genetic changes altered the

teeth of Tertiary horses in ways to make the latter "preadapted" to

grazing instead of browsing (page 305). Man's erect posture made

his hands "preadapted" towards performance of delicate manual opera-

tions (Chapter 13).

The difficulty is chiefly a verbal one. When some ancient anthro-

poids made the first tentative attempts to walk erect, not even the

shrewdest evolutionist could have predicted that the fore paws of

these anthropoids would some day handle surgical instruments used

for delicate operations. When these anthropoids were increasing their

brains in response to the need for more cunning in escaping enemies

it was quite unpredictable that the brains of their remote descendants

would eventually develop the ideas of Plato or of Darwin. In other

words, we must have the benefit of hindsight to decide whether a

given change was preadaptive for anything. "Preadaptation" is a

meaningless notion if it is made different from "adaptation."

Progress in Evolution. We have seen in Chapter 10 that according

to Aristotle all living beings range in a single series from the least

perfect to the most perfect. Lamarck thought that this single ladder

of perfection reflected evolution, and this unfortunate mistake was

largely responsible for the rejection of Lamarck's whole theory. The

belief in progress, however, was too dear to the nineteenth century

mentality, and Darwin's theory was promptly interpreted to mean

that "no good thing was ever lost and that no lost thing was any longer

any good" (Barzun). Indeed, natural selection generally does main-

tain or improve the adaptedness of a living species to its environment.

The situation, however, is complex. In the first place, there is no

single line of evolution but very many different lines. Furthermore,

every kind of organism occupies its own adaptive niche, to which it

is fitted by its body structure and by its mode of life. The mighty

lion may be the king of beasts, but it cannot fly through the air like

a bird, nor live in water like a fish, nor subsist on simple organic com-

pounds like bacteria, nor utilize the sunlight as a source of energy

like green plants. By and large, every species is superior to all otljers

in its own adaptive niche, for if another species were superior it would

drive the first out. But it is ridiculous to conclude on this basis that

there has been no progress in evolution, and that man is not a more

perfect organism than a worm or an amoeba. There are several

possible criteria of progress and perfection, and we must make it

clear which one is being used.

Progress in Evolution 371

Evolution has brought about a tremendous increase in the range of

environments in which life is possible, in the diversity of living crea-

tures, and in the total mass of living matter. We do not know what

the primordial life was like, but it is certain that it could exist only

in some very restricted range of environments. The fossil record of

the early Paleozoic consists of only water-dwelling creatures, while

land was apparently lifeless. At present life extends almost every-

where on the earth's surface, including such apparently inhospitable

places (from the human standpoint) as deserts, high mountains, and

subpolar regions. Some forms of life seem so grotesque in appearance

(Figure 14.2), or live in such strange places, that we naively wonder

why such creatures ever appeared. This is another instance of the

opportunism of evolution: whatever can perpetuate itself in any

accessible environment does so. The appearance and spread of a

new species constitute progress, whereas extinction is regression.

Although evolution has been, in this sense, progressive on the whole,

the living matter (biosphere) is still, on a planetary scale, merely a

thin film at the boundary of the earth's crust (lithosphere) and its

gaseous envelope (atmosphere).

Another kind of progress is the increasing complexity of organiza-

tion of living beings. Simplest plant viruses consist apparently of a

single chemical substance—nucleoprotein. Bacteriophages and, even

more so, bacteria and unicellular organisms are immensely more com-

plex, since their bodies are structured systems involving numerous

chemicals. Then follow multicellular organisms; these again range in

complexity from such relatively "simple" things as Pandorina (Figure

1.2), which are merely colonies of semi-independent cells, to very

complex systems of cells, such as the "higher" animals and man. In

this respect, too, evolution has been on the whole progressive, although

not always so. For example, parasitic organisms, such as Sacculina

(Figure 10.6), are often much simpler in structure than their free-

living relatives. There even exists a view that the viruses may be

degenerate descendants of more complex bacteria-like organisms.

Furthermore, it is not just a love of complexity which makes evolu-

tion produce ever more intricately organized living bodies. Although

there is no strict relationship between the two things, more complex

organization by and large gives a greatei autonomy of the organism

from the environment. As pointed out especially by Schmalhausen

(1949), evolution has tended to improve the homeostatic adjustments,

so that the processes of life can go on despite environmental variations.

The transition from the "cold-blooded" reptiles to the "warm-blooded"

Man, the Pinnacle of Evolution 373

ably certain that this probability is, in general, improved in the higher,

compared to the lower, organisms.

Although progress is not necessarily the same thing as success, it is

a fact that those groups of organisms which on other grounds may be

considered most progressive are likely to become also abundant, di-

versified, and "dominant." The succession of the "ages" of amphibians,

reptiles, mammals, and man attested by the fossil record (Chapter 12)

is a case in point. This succession corresponds to what is regarded as

progressive evolution among vertebrate animals. A similar reward for

evolutionary progress was the accession to dominance of the "higher"

flowering plants at the end of the Mesozoic era. On the other hand, a

failure to "progress" does not always result in the organism's becoming

rare and extinct. Among the "living fossils" shown in Figure 12.3 at

least the horseshoe crab (Limulus) and the opossum, and probably

also Lingula, are common and prosperous species, apparently in full

possession of their respective adaptive niches.

Man, the Pinnacle of Evolution. One of the cheapest ways to

gratify our ego is to consider ourselves superior to others. For this

reason, the opinion that man stands on the topmost rung of the ladder

of progress must be carefully scrutinized. It happens, however, that

by all sensible criteria of progress man is superior to other creatures.

Mere growth in numbers may not be an unmitigated blessing, as

man is finding to his discomfiture in the overpopulated countries of

southern and eastern Asia, but mankind has increased manyfold since

the invention of agriculture, and has occupied most of the habitable

.surface of the globe (Chapter 13). The case is less clear in com-

plexity of the organization of the body, since there is no way to

measure this complexity precisely. We can say, however, that the

vertebrates in general, and among them the mammals, possess re-

markably elaborate organic systems. Man belongs to the class of

mammals. There can be not the slightest doubt that man is now the

dominant species; with the development of biological technology all

other species will exist only on man's sufferance.

The conclusive evidence of man's superior position is that he, and

he alone, has evolved the genotype which enables him to develop and

maintain culture. As pointed out in Chapter 13, the transmission of

the cultural inheritance is superimposed on biological heredity, but

the former is a vastly more efficient process than the latter. Biological

heredity is handed down only from parents to children; culture can

be transmitted to anybody. Acquisition and transmission of culture

have conferred upon man as a species an unprecedented degree of

374 Chance, Guidance, and Freedom in Evolution

fitness in the sense of Thoday's definition (see above). Unless man-

kind chooses to destroy itself by atomic explosions or similar means,

it is more likely to endure than any other creature. Man's biological

pinnacle is a solitary eminence; no other species can aspire to dis-

pute it.

Biology gives no warrant for the belief that man was preformed in

the primordial life, or that the evolution of life as a whole had as its

purpose the production of man. Evolution does not strive to accom-

plish any particular purpose or to reach any specific goal except the

preservation of life itself. Evolution did not happen according to a

predetermined plan. Nevertheless, when man contemplates the whole

perspective of the evolution of the Cosmos, he can see that the origin

of mankind was one of the outstanding events in the history of creation

by evolution. This event is represented symbolically in Michelangelo's

beautiful fresco reproduced as the frontispiece of this book.

Considered biologically, man arose because of the action of the very

same forces which bring about the evolution of all other organisms.

Natural selection responded to the challenge of environmental oppor-

tunity, and compounded an adaptively highly successful genotype

from genetic elements contributed ultimately by mutation. In a sense,

the origin of man was a lucky accident. Evolution is not repeatable,

because slight differences either in the environment or in the genetic

materials might have resulted in something different from man. But

in another sense man was not accidental. Natural selection is, in

Fisher's words, "a mechanism for generating an exceedingly high de-

gree of improbability." In other words, selection creates gene com-

binations which would be almost infinitely unlikely without it. Laplace

was right, but only in part. Perhaps a "vast intelligence" could dis-

cern in primordial matter the distant coming of man; but this intel-

ligence had to be divine, not human.

Genetically Determined Educability. It has been pointed out in

Chapter 13 that man, like other biological species, became genetically

differentiated into races. This differentiation occurred in response to

the differences in the physical environments of the different countries

which man inhabits. But as time went on, man's physical environ-

ments became more uniform or more easily controllable. The reverse

was true of cultural environments. Here the diversity is steadily

increasing. Man is inseparable from his culture, and cultural environ-

ments can be dealt with only by learning to choose among many

possible courses of action the one appropriate under the circumstances.

Biological evolution has produced the genetic endowment which has

Equality of Men 375

made culture and freedom of choice possible. But from then on,

human evolution has become in part a new and unprecedented kind

of evolution—evolution of culture and of freedom. This certainly

does not mean that the biological evolution of man has come to a

halt, as some writers like to suppose. The two kinds of evolution,

biological and cultural, are combined in a new and unique process

which is human evolution.

Human behavior is conditioned by education in a wide sense of this

word, which is the sum of a person's experiences. The conditioning

occurs from infancy on—coming from parents, playmates, neighbors,

teachers, companions, friends, enemies, and also from books, news-

papers, and all other means of communication. Of course the out-

come of the conditioning depends upon the person conditioned. Edu-

cation is useless without a brain to absorb it, and an educable brain

presupposes a human genetic endowment. The biological uniqueness

of the human genotype lies in the fact that it permits a greater degree

of educability than the genes of any other biological species.

Equality of Men. The problem of the equality of men is a topic

of discussion which often generates more heat than light. Evolution-

ary biology should help at least to state the problem correctly. Equal-

ity and inequality of men are religious, ethical, and legal, not biological

concepts. Men may be equal before God and before the law without

being biologically alike. Indeed, men are not biologically alike; no

two men, identical twins excepted, have the same genotype. To some

people this fact is emotionally repugnant, a repugnance which can be

due only to misunderstanding of the meaning of heredity. Heredity

of mental, emotional, and personality development is mostly condi-

tioning, not destiny. Excepting genetically controlled mental diseases,

"normal" human genotypes permit a great latitude of intellectual and

emotional developments.

Whether the genetic endowment makes some people superior and

others inferior is another story. This question has no meaning unless

the basis of the value judgment is made explicit. Superior or inferior

for what? Most people will agree that a congenital idiot who is

incapable of performing ordinary work and unable to take care of

himself is not a useful member of any society. But is a football player

superior or inferior to a chess player? How could we compare the

values of a scientist and an artist? Of a thinker and a man of action?

Of a farmer and an industrial worker? Clearly, mankind needs all

of these and many other kinds of people. And, fortunately, most

human individuals can be trained to perform competently many of

376 Chance, Guidance, and Freedom in Evolution

the functions that are needed in human society, although some indi-

viduals may be more successful in some functions than in others.

Here one must beware of a specious argument which is the more

misleading because it is so plausible. The argument runs about as

follows. Since we know that races and breeds of wild and domestic

animals often differ in genetically conditioned behavior, how can it

be that human races are an exception? The answer is simple. The

race horse and the draft horse, or the fox terrier, the dachshund, and

Great Dane, differ in their temperaments, and these differences are,

to a large extent, genetic. But these differences are there because they

have been built into these breeds by the artificial selection which

fashioned them to serve different needs or whims of their owners.

Certainly a race horse with a temperament of a draft horse would be

a failure, just as a draft horse with a temperament of a race horsr

would be. Now human races, no less than human individuals, suffer

frequent and often drastic changes of fortune. The adaptive advan-

tage of educability is the one constant in human evolution since the

beginning of its cultural, or truly human, phase. Genetic differences

between races are, then, secondary to those differences rooted in

cultural heredity.

Evolutionary Ethics. Biological evolution has contrived human

genotypes which make man, in the words of Thomas Jefferson, "formed

for society, and endowed by nature with those dispositions which fit

him for society." Among these "dispositions," surely one of the most

important is the ability to distinguish between right and wrong. It is

man's moral sense which makes him truly human. Where does this

moral sense come from? All religions claim either that ethics are

based on supernatural revelation or that the ability to discriminate

between good and evil has been implanted in the human soul by

God. On the other hand, according to Chauncey Leake: "The proba-

bility of survival of a relationship between individual humans or

groups of humans increases with the extent to which that relationship

is mutually satisfying." May, then, man's ethical sense be a product

of biological evolution by natural selection? Or, perhaps, the re-

ligious and the evolutionary explanations are the two sides of the same

coin?

Flushed by the tremendous successes of natural science in their

time, some nineteenth century scientists felt that no problem could be

insoluble in mechanistic terms. In his Principles of Ethics (1892),

Herbert Spencer (1820-1903) attempted to show that ethics are "part

and parcel" of biological evolution. Good conduct makes for pleas-

Evolutionary Ethics 377

urable life in society; wrong actions are socially disruptive. Natural

selection has encouraged the spread in human populations of those

moral qualities which were useful for the preservation of the life of

individual members of society, and hence for the maintenance of

society as a whole. Thomas H. Huxley (1825-1895) entertained, for

a time, ideas like Spencer's; however, in his celebrated lecture given

in 1893 he pointed out quite clearly the weaknesses of these ideas.

According to him, "Cosmic evolution may teach us how the good and

the evil tendencies of man may have come about; but, in itself, it is

incompetent to furnish any better reason why what we call good is

preferable to what we call evil than we had before." And further:

"The practice of that which is ethically best—what we call goodness or

virtue—involves a course of conduct which, in all respects, is opposed

to that which leads to success in the cosmic struggle for existence."

Julian Huxley, a grandson of T. H. Huxley, is the most active modern

exponent of "evolutionary ethics." Although philosophers and theo-

logians usually believe that the validity of an ethical code must rest

on divine sanction, J. Huxley proposes "a morality of evolutionary

direction." Although not a believer in orthogenesis, J. Huxley thinks

that he can discern a direction in biological and in human evolution,

and that this direction "consists basically of three factors—increase in

control over the environment, increase in independence of the environ-

ment, and the capacity to continue further evolution in the same

progressive direction." If so, then "anything which permits or pro-

motes open development is right, anything which restricts or frustrates

development is wrong."

This is certainly an interesting idea, but it may be questioned

whether it meets adequately the above-quoted objection of T. H.

Huxley against evolutionary ethics. Suppose that the evolutionary

direction which our life has followed until now can be known beyond

reasonable doubt. Would this necessarily mean that this direction is

good and that we ought to help its continuance? Evolution has pro-

duced a mind capable of knowing that it has evolved and that it can

evolve farther. Can't this mind also scrutinize the wisdom of the

process which gave birth to it? But such a scrutiny can be based only

on criteria of wisdom derived from sources other than the evolution-

ary process itself. Man knows not only that he has evolved and

continues to evolve, but that he is also in the process of learning how

he may promote his own evolution in the direction of his choice. But

what should be the basis of the choice? As G. G. Simpson rightly

wrote: "It is futile to search for an absolute ethical criterion retro-

378 Chance, Guidance, and Freedom in Evolution

actively in what occurred before ethics themselves evolved. The best

human ethical standard must be relative and particular to man and

is to be sought rather in the new evolution, peculiar to man, than in

the old, universal to all organisms." Evolutionary ethics have not been

formulated yet, and one may reasonably doubt that they can be made

scientifically convincing or aesthetically satisfying.

Epilogue. The evolution of the Cosmos has not been everlastingly

uniform. Some important "dates" can be discerned which mark pe-

riods of titanic stress and crisis, leading to events of transcendental

significance. The first date is, of course, the beginning of the universe

itself. According to the estimates which physicists and geologists are

able to make, this beginning occurred some 5 billion years ago. Some

2 billion years ago life appeared there, and biological evolution was

inaugurated. As far as anybody knows for sure, this happened, in the

whole wide universe, only on a tiny speck of dust which is our earth.

Biological evolution created some millions of species of organisms,

which explored various possibilities of living. Most of these species

eventually became stranded in the blind alleys of opportunistic con-

formity to the favorable situations in the environment which proved

only fleeting and temporary. But some organisms made evolutionary

"inventions," which permitted them to spread, multiply, and, some of

them, to become dominant. Half a million to a million years ago, one

species made an evolutionary "discovery" of unparalleled significance;

it became capable of extra-biological transmission of acquired and

learned experience. This species became human, and opened up a

new, cultural or human, evolution. About two thousand years ago

this species had advanced far enough to be able to receive the Sermon

on the Mount. The development of science, at first slow and hesitant,

but during the last two centuries rapidly accelerating, is enabling man

gradually to acquire a better understanding of himself and of his

environment on earth and in the Cosmos. Julian Huxley thinks that

man "finds himself in the unexpected position of business manager for

the cosmic process of evolution." This judgment may be a premature

one, but man does struggle against the bounds of his nature, and this

struggle makes his existence worth while, and contains a hope of a

noble future.

Suggestions for Further Reading

Berg, L. S. 1926. Nomogenesis or Evolution Determined by Law. Constable,

London.

Suggestions for Further Reading 379

This is perhaps the most thoughtful book ever written by a believer in evolu-

tion by autogenesis.

Noiiy, Lecomte du. 1947. Human Destiny. Longmans, Green, London. (Also

available in a paper-bound edition of The New American Library.)

This is a popular presentation of the finalist point of view. The book contains

numerous and important factual errors in its presentation of biological and geo-

logical information. A point of view related to finalism is presented also in the

following article:

Osborn, H. F. 1934. Aristogenesis, the creative principle in the origin of species.

American Naturalist, Volume 68, pages 193-235.

For a critique of the autogenetic and finalist theories of evolution, see particu-

larly G. G. Simpson's books, quoted among the suggested readings in Chapter 12.

Concerning the evolution of man and of man's unique abilities, see the suggested

readings in Chapter 13.

The problem of the so-called scientific ethics is discussed particularly in the

following books:

Huxley, T. H., and Huxley, J. 1947. Touchstone for Ethics. Harper, New York

and London.

Huxley, J. 1941. Man Stands Alone. Harper, New York.

Huxley, J. 1953. Evolution in Action. Harper, New York.

Leake, Chauncey D., and Romanell, P. 1950. Can We Agree? University of

Texas Press, Austin.

Waddington, C. H. 1942. Science and Ethics. Allen & Unwin, London.

Index

Abel, 274

Aberrations, chromosomal, 63-68, 83

in Drosophila, 143-148, 176, 219

Abiogenesis, 16

Achillea, 154-157, 163

Acorn worm, 244-246

Acquired characters, 77-80, 89, 115

Adaptation, 12, 13, 35, 73, 91, 110,

154-155

and diversity, 162, 167

and modification, 75-77

and mutation, 106-107

and preadaptation, 369, 370

chemical, 244

complex, 368, 369

for cross-fertilization, 259

in bacteria, 92-100

in man, 5, 334, 335

origin by selection, 14, 19, 91, 102,

109-133, 158, 159, 290

Adaptive niche, 370

Adaptive plasticity, 345

Adaptive radiation, 291-293, 295, 296,

307, 314, 315, 326, 366

Adaptive value, 119, 120, 122, 123, 147,

148, 177

Adenosine triphosphate, 243-246

Adrenaline, 274, 275

Age of Amphibians, 289-291, 373

Age of Mammals, 293-296, 373; see

also Mammals

Age of Reptiles, 373; see also Reptiles

Albertus Magnus, 166

Albinism, 74, 76, 140, 248

in man, 31, 33, 40, 42

Algae, 9, 12, 91, 254-256

Alice, 113, 133, 341, 356

Alleles, see Gene alleles

Allison, 143, 219

Allium, see Onion

Allomctric growth, 346

Allopatric, 137, 138, 160, 163, 167-169,

179, 183-188, 345

Allopolyploids, 168, 202, 204, 208

Alloterraploids, 204-206

Amadon, 314, 318

Amaurotic idiocy, 32, 107, 140

Amino acids, 7, 11, 18, 19, 83

Amoeba, 12

Amphidasys, 104

Amphioxus, 369

Amputation, effects of, 78

Analogy, 231, 233-235, 238, 247, 249,

306, 308

Anaphase, 46-48, 50

Anatomy, comparative, 227-232

Anaximander, 3, 357

Ancon sheep, 81

Anderson, Edgar, 127, 133, 213

Anderson, E. H., 98

Andrewartha, 133, 163

Anemia, 32, 219

Aneuploid, 64, 66, 82

Animalculists, 222

Anthropoid apes, see Apes

Anthropology, 135, 137, 159, 160-162,

346-354; see also Man

Anthropomorphism, 343

Antibiotics, 96, 98

382

Index

Aquinas, St. Thomas, 357

Archaeopteryx, 110

Arctostaphylos, 174

Arginine phosphate, 245, 246

Aristogenesis, 360

Aristotle, 16, 44, 223, 224, 227, 233,

253, 256, 357, 370

Art, prehistoric, 6

Artificial insemination, 173

Artificial new species, 207

Artificial selection, 125, 179, 191, 376;

see also Selection

Ascherson, 209

Ascidians, 7; see also Ciona

Asexual reproduction, 115, 123-125,

168, 255

Asexual species, 188, 189

selection in, 126

Ashley Montagu, 113, 133, 327, 341,

355

Asses, 169-172, 177, 182, 193, 196, 197,

304, 305

Assimilation, 7, 11, 19

Atoms, 2, 4, 319

Auerbach, 87

Australian fauna, 315, 316

Australopithecus, 5, 326-329, 333, 334,

336, 338, 362

Autogenesis, 358-362, 365, 368

Autopolyploids, see Polyploids

Autosomes, 262-265

Aye-aye, 324

Babcock, 47

Baboon, 246

Bacci, 271

Backcross, 54

Bacon, 16

Bacteria, 14, 16, 17, 371

adaptation of, 92-100

genes in, 33

mutation in, 83, 87, 92-100

reproduction of, 23, 111

sex in, 33

virulence of, 99

Bacteriophages, 15, 33, 92-95, 106, 371

Baer, von, 223, 235, 238

Balanced polymorphism, see Poly-

morphism

Balanoglossus, 244-246

Balbiani, 46

Baldwin, 245, 252

Baltzer, 173, 271

Banteng, 174

Bark beetles, 103

Barley, 124, 166, 194, 196, 217

Barnacle, 239-240

Barzun, 370

Basic plans, see Types

Bateson, 43

Bats, 231, 232, 292, 307

Bauer, 46

Bayliss, 274

Beadle, 83, 89, 249

Beans, 113, 114, 119, 194, 196, 217

Beasley, 207

Bedbugs, 103

Beer, de, 252

Bees, 229, 230, 269-270, 341

Index

383

Blakeslee, 174, 265

Blending inheritance, see Heredity

Blood, adaptation of, 13

groups, 125, 130, 136, 138, 139, 149-

152, 188; see also Race and Man

theory of heredity, see Heredity

transfusion of, 79

Blue grass, 64

Blum, 18, 21

Boarmia, 105

Bonellia, 270-271

Bonnet, 224, 225, 231

Bonnier, 277

Boveri, 46, 53

Bowater, 104

Bower birds, 279, 280

Box theory, 222

Boyd, 164, 355

Brachiopods, 296

Brachycephalic, 42, 153

Brain, 334-337, 361, 370-372, 375

Brassica, see Cabbage

Braun, 108

Breakdown, hybrid, 172, 177-179

Breeds, 42, 138, 193, 194, 376; see also

Dog and Horse

Bridges, 55, 63, 262, 263, 273, 274

Brncic, 178

Broom, 326

Brown, 45

Browsing, 303-305, 370

Buchsbaum, 251

Budgerigar, 192, 193

Buffalo, 174, 193

Buffo, see Toad

Buffon, 77

Burks, 90

Burnet, 92

Burrows, 283

Burton, 283

Butenandt, 275

Buthus, 282

Butler, 79, 113

Biitschli, 46

Cabbage, hybrids with radish, 205-207

Calvinists, 223

Camarhynchus, 341

Cambrian life, 2, 288, 289

Camels, 311

Camouflage, 105

Candolle, de, 194, 242

Care of the young, 294, 295

Carnivores, 309

Carpocapsa, 104

Castes in social insects, 269, 270, 343

Castration, 275-276, 281

Catcheside, 108

Catechol, 100

Cats, 193, 248, 311

Cattell, 90

Cattle, domestication of, 193

hybridization of, 127, 174

milk production in, 122

serology of, 247

twins in, 248

Caullery, 260, 283

Cell theory, 45

Centers of origin of domestic species,

384

Index

Chromosomes, disjunction of, 46, 63

genetic maps of, 55-58, 68

in Crepis, 47

in man, 59

in salivary glands, 46

numbers of, 46, 63

origin of, 70

pairing of, 52

Ciona, 7, 259

Classification, and types, 134, 135

artificial, 232, 233

natural, 232-234, 247

Clausen, 154, 155, 163

Clements, 75

Clines, 149, 161; see also Race

Clones, 115, 123-125, 131, 136, 155,

189, 255

Coadaptarion, 177

Coat colors in mammals, 37—40

Coccids, 104

Coccyx, 241, 242

Cockroach, 103, 229, 230

Codling moth, 104

Coe, 260, 261

Coelacanths, 296, 297

Coffee, 196, 207

Colbert, 110, 287, 294, 307, 318, 336

Colchicine, 65

Colias, 272

Colletotrichum, 100

Colon bacteria, 23, 92-95, 96, 97, 106

Colonial organisms, 9, 10

Colonies, isolated, 129, 130

Color blindness, 62, 249

Coloration, protective, 105

Columbines, 175

Competition, 111-113; see also Natural

selection

Complex adaptations, 368, 369

Condylarths, 298, 360

Consanguinity, see Inbreeding

Continental drift, see Drift

Continuity of life, 16, 17

Convergent evolution, 232, 247, 291

Coon, 164, 333, 347-350, 354, 355

Cooperation, 112, 113, 341-344

Cope, 305

Corals, 9

Corn, chromosomal aberrations in, 68

chromosomes of, 49-51, 54

cross pollination of, 215

double crossed, 215, 216

genetic maps in, 55

hybrid, 142, 214-217

inbreeding in, 216, 217, 219

mutations in, 85, 86

origin of, 192, 196, 208-214

primitive, 210-212

relatives of, 208-210, 260

selection in, 125-126

sex in, 260

varieties of, 213-217

Corpus luteum, 275

Con-ens, 25

Cosmic evolution, 1-5

Cosmic rays, 88

Costa Lima, 372

Cotton, domestication of, 194, 196, 201,

Index

385

da Cunha, 219, 220

Curtis, 93

Cuvier, 135, 224, 225, 227, 233, 234

Cyclops, 238-240

Cyclostomata, 236-238

Cytogenetics, 45

Cytology, 45, 53

Cytoplasmic heredity, 88, 89, 267

Dacque, 359

Darlington, 108, 251

Dart, 5, 326, 328

Darwin, 13, 23, 25, 77-79, 82, 91, 108-

113, 115, 117, 133, 165-168, 181-

183, 188, 190, 192, 194, 215, 217,

227, 233, 236, 242, 250, 253, 277,

278, 285, 288, 311, 315, 319, 355,

357, 358, 368, 370

Darwinian fitness, see Adaptive value

Darwin's finches, 311-313

Datura, 174

Daubentonia, 324

Dauvillier, 18

Davenport, 40

DDT, 102, 103, 106

Deer Mice, 174

Deficiency, chromosomal, 65-69, 81

Delbruck, 92

Deleterious mutants, see Mutation —

Deletion, see Deficiency

Demerec, 87, 92, 94, 96, 98

Democracy, 162

Derjugin, 245

Descartes, 357

Desguin, 18

Desoxyribose nucleic acid, 8, 46

Determinants, 250; see also Weismann

Development, individual, 35-38, 44, 77,

223, 242, 250, 369

Diabetes, 76

Diakinesis, 52

Dice, 174

Didelphis, see Opossum

Diffusion of genes, see Hybridization

Dinosaurs, '291-493

Dioecious plants, 257, 258

Diploid chromosome set, 51

Diplotene, 50, 52

Diseases, hereditary, 32, 36, 76, 83, 85,

128-130, 139-143, 219; see also

Man

Disjunction of chromosomes, 46, 63

Dispersal, sweepstakes, 313, 317; see

also Migration

Divergence, evolutionary, 167

Dobzhansky, 43, 65, 70, 90, 133, 145,

163

Dog, 135, 169, 192, 193, 194, 247, 248,

335, 337, 338, 364

Doisy, 275

Dolichocephalic, 153

Domestic animals and plants, 122, 127,

138, 184, 191-221, 284; see aho

Barley, Cats, Cattle, Corn, Cotton,

Dog, Horse, Poultry, Rye, and

Wheat

Dominance, 25, 29

Drepaniids, 314

Drift, continental, 317

Index

Environment, challenge of, 122, 123,

131, 132, 363, 374

human, 345; see also Culture

Environmental opportunity, 363

Enzymes, 7, 243-246, 249, 251

Eohippus, 298-300, 302, 303, 305, 360

Epidemics, 99-102

Epigenesis, 222, 223, 235, 242, 250, 251

Epiloia, 85

Epinephrin, 274-276

Epling, 163

Equality of men, 162, 375, 376

Equidae, see Horse

Equilibrium, genetic, 143, 144, 146, 147

Equus, see Horse

Escherichia, see Colon bacteria

Estrogen, 275, 276

Ethics, 376-378

Ethiopian fauna, 315, 316

Ethnic groups, 160; see also Man, races

of

Drosophtia pseudoobscura, 47, 68, 144,

146-148, 152, 159, 163, 170-172,

176, 178, 179, 188, 266, 279

repleta, 47

simulans, 248

subobscura, 47, 279

virilis, 172

wiOistoni, 47, 142

Dryopithecus, 326

Duckbill, 234, 293

Duoks, 174

Dunaliella, 254

Dunbar, 318

Dunn, 43, 65, 70, 83, 108, 133

Du Noiiy, 359

Duplications, 65-67, 69, 208

Dwarfs, 32, 85

Earthworms, 257

Echinoderms, 173, 244-246

Ecological isolation, see Isolating mech-

anisms

Ecological niche, 178, 305, 311, 314,

367

Educability of man, 374-376; see also

Culture

Egg cells, 10, 11, 45, 52

Eggs, numbers of, 111

Ehringsdorf skull, 332

Eimer, 360

Electron microscope, 14, 15, 33, 92, 95

Elephants, 363

Elk, Irish, 360, 361, 364

Embryo sacs, 52

Embryology, 235-238, 241; see also De-

velopment

Emerson, 356

Empedocles, 111, 357

Empidae, 279

Endemic, 311, 313

Endocrines, 274-276, 344, 368

Environment, and adaptation, 14, 153,

290

and heredity, 11, 12, 13, 19, 23, 69,

72-92

and mutation, 80, 86-89, 106

and sex, 268-272

Eucalypts, 315, 316

Euchlaena, 208-210

Index

387

Extinction, 289, 366-368

Eye colors in man, 31, 33

Fabre, 279, 344

Fabricius, 235

Factor of safety, 269

Falls, 139

Fano, 92

Faunal regions, 315-317

Fecundity, 119

Federley, 206

Femaleness, 263, 267, 271

Fertilization, 44, 45, 52

interspecific, 173

Filter flies, 103

Finalism, 20, 358-362, 369; see also

Vitalism

Finches, 312, 313

Fisher, 120, 374

Fitness, Darwinian, see Adaptive value

Flax, 174, 196

Fleming, 46

Flint, 330

Floral regions, 315-317

Florkin, 252

Fluctuating variability, 82; see also

Polygenes

Fontechevade skull, 332

Ford, 105, 219

Form, organic, 222-252

Fossil evidence of evolution, 284-318,

328-337, 360-368, 371

Fossil man, 328-337

primates, 325, 326

Fossils, Cambrian, see Cambrian life

Fox, 158

Fraternal twins, see Twins

Freedom in evolution, 357-378

Freemartin, 277-278

Frequency of mutation, 84-88; see also

Mutation

Frisch, von, 341

Frogs, 178, 179, 184, 238, 258, 290

Function, organic, 222-252

Galapagos Islands, 311-313, 316

Galeopsis, 207, 208

Galton, 25, 79

Gama grass, 209, 210

Gametes, 49, 51, 52, 254

Gamow, 21

Gam, 164, 347-350, 354, 355

Gates, 43

Gemmules, 79, 80, 250

Gene, 206

Gene alleles, 33

arrangement in species, 143, 206

balance, 68

correlated with chromosomes, 53-63

defective, 32; see also Diseases and

Lethals

definition of, 25

destruction of, 80

dosage of, 65-67

effects on development, 35-38

frequencies, 115-121

homology, 247-248

in siblings, 32-35

independent assortment of, 27-34, 54,

55

388

Index

Heterozygote, 27

Hevea, 192

Hiesey, 154-157, 163

Hipparion, 303-305

Historical record, see Fossil evidence of

evolution

Hofmeister, 45

Hogben, 90

Holarctic fauna, 315

Holmes, 3

Hologenesis, 359

Homeosis, 83, 84

Homeostasis, 13, 14, 76, 133, 293, 345,

371

Goldschmidt, 84, 165, 266, 267

Gorilla, see Apes

Gossypium, see Cotton

Gowen, 99, 100, 108, 221

Graafian follicles, 275

Gradients, geographic, 149, 161

Grant, 175

Grasshoppers, 50, 56, 229

Grazing, see Horse

Gregory, 252, 318

Grew, 235

Ground plans, see Types

Growth, allometric, 364

Gulls, 185, 186

Guthrie, 93

Gynandromorphs, 272-274

Gypsy moth, 266, 267

Hadom, 83

Haeckel, 234, 236, 241, 242

Haldane, 85, 90, 120, 139, 243, 252

Haploid, 49, 50, 52, 270

Hardy-Weinberg law, 117-120, 136

Hares, 173-247

Harlan, 124

Harland, 204, 205, 221, 248, 251

Harms, 281

Hartmann, 270, 283

Harvey, 16, 235 ,

Haskins, 356

Hawaii, fauna of, 312-314

Heitz, 46

Hemophilia, 62, 85, 107

Hereditary diseases, see Diseases

Heredity, and environment, see Envi-

ronment

biological, 4-5; see also Gene

blending, see Polygenes

blood theory of, 24, 25, 35, 42, 117

chromosomal theory of, 62

cultural, see Culture

cytoplasmic, 88-89

evolution of, 69, 70

Hermaphrodites, 119, 257-261, 271

Hershey, 95

Hertwig, 45

Heterosis, 142-144, 148, 176, 215-220

balanced, 218-220

mutational, 218-220

Homo, see Man

Homology, 228-235, 238, 308

biochemical, 243-246

of genes, 247-249

serial, 227

special, 227

Index

Hybridization, intergeneric, 174

interspecific, 172, 175, 178, 179, 197,

206, 208

introgressive, see Introgression

relation to selection, 33, 126, 131

remote, 173

Hydrocyanic gas, 104

Hyracotherium, see Eohippus

Ideas, Platonic, 134, 225

Identical twins, see Twins

Idiocy, see Amaurotic idiocy

Immunology, 246, 247

Inbreeding, 119, 177, 215, 217, 218,

219, 258

Incense cedar, 317

Independent assortment, see Genes

Indomalayan fauna, 315, 316

Indus civilization, 201

Industrial melanism, 104, 105

Infection, 13; see also Epidemics and

Viruses

Infusoria, 89

Insecticides, 102-104, 341-344

Insectivores, 295, 325

Insects, mouth parts of, 229, 230

Insemination reaction, 173

Instinct, 339, 344

Insulin, 275

Intelligence, 335, 337; see also Culture

Intermarriage, 138

Internal secretion, 274-276

Interphase, 49

Intersexes, 261-269, 276, 277; see also

Drosophila

Introgression, 127, 174, 179, 214; see

also Hybridization, interspecific

Inventions, evolutionary, see Evolution-

ary inventions

Inversions, 65-69, 206, 208; see also

Drosophila, inversions in

Inviability of hybrids, see Hybrid invia-

bility

Iris, 127

Irish elk, 360, 361, 364

Irritability, 7, 12

Island continent, 307

Islands, continental, 311-313

oceanic, 311-314, 316

Isolates, 137

Isolating mechanisms, 169, 172-174,

179-180, 201; see also Hybrid

breakdown

genetic, 172

ecological, 170, 187, 188

ethological, 171

geographic, 168, 169, 188

habitat, 170-174

reproductive, 168-170, 172-179, 181,

182, 184-187, 206

sexual, 171-173, 179, 278

Isotopes, 12, 287

Ivanovsky, 14

Ives, 88

Java Man, 328, 329, 331, 335-337

Jefferson, 376

Jimson weed, see Datura

Johanssen, 25, 73, 74, 113-115, 123,

124, 130, 250

Jones, 215, 260

390

Index

Lamprey, 236-238, 288

Lancelet, 369

Land bridges, 308

Landauer, 83

Lang, 126

Language, see Culture

Laplace, 358, 374

Larus, 185, 186

Larvae of crustaceans, 238-240

Lascaux caves, 6

Latimeria, 296, 297

Lavoisier, 243

Law of independent assortment, 27-29

Law of segregation, 25-27

Le Gros Clark, 330-338

Lead, 287, 359

Leake, 376, 379

Leakey, 326

Learned behavior, see Culture

Lederberg, 33, 108

Leeuwenhoeck, 45, 222, 261

Leibnitz, 223, 224

Lemurs, 322, 324, 325

Leonardo da Vinci, 44

Lepas, 239, 240

Leporids, 173

Lerche, 254

Lerner, 43, 133, 221

Lethals, 82, 83, 86, 105, 106, 139, 140-

142; see also Diseases and Man

Li, 63, 133

Libocedrus, 317

Lice, 103

Life, continuity of, 16, 17

nature of, 1-22; see also Self-repro-

duction

origin of, 1-22, 123, 288

Lillie, 277

Limnopithecus, 326

Limulus, 296, 297, 373

Lindquist, 103

Linear arrangement of genes, 54-60

Lingula, 296-298, 373

Linkage, 53-58

Linnaeus, 135, 166, 167, 181, 189, 232-

234, 320

Litopterns, 309-311

Locy, 251

Lorenz, 356

Lucretius, 111, 357

Luria, 92-95

Lymantria, 266, 267

Lysenko, 16, 79, 89, 166

Macroevolution, 166, 191

Macrospores, 52

Maize, see Corn

Malaria, 143

Maleness, 263, 267, 271

Malpighi, 223, 235

Malthus, 111

Mammals, adaptive radiation of, 292

age of, 293-296

embryos of, 235-238

evolution of, 4, 13, 295, 365, 366

in Americas, 308-311

origin of, 363

see also Placenta! mammals and Mar-

supials

Index

391

Marsh, 305

Marsupials, 293, 306, 308-310, 315, 367

Martini, 124

Mason bee, 344

Mather, 43

Matthew, 305, 311, 318

Maupertuis, 79, 357

Mayr, 158, 161, 163, 168, 185, 318

McClung, 55

McCown, 332

McDougaU, 78

Measles, 76

Mechanism, 19, 20

Meckel, 235

Megaloceros, 360, 361, 364, 368

Meiosis, 49-52, 55, 255

in cotton, 203-204

in Drosophila, 57

in man, 59

in species hybrids, 206, 208

origin of, 70

Meiotic divisions, 50, 52, 56, 59

Melandrium, 265

Melanism industrial, 104-105

Melipona, 270

Membracids, 372

Memory, 79

Mendel, 24-29, 36, 43, 45, 82, 253

Mendelian populations, 134-165, 177,

182, 184, 333

as unit of selection, 147

definition of, 115

equilibrium in, 147

human, 137; see also Race

Mendel's laws, 27-30, 33, 34, 54

Merychippus, 304

Mesohippus, 301, 302

Mesopithecus, 336

Metabolism, 12, 20

Metanephros, 236-238

Metaphase, 46-48, 50

Microevolution, 90-108, 165, 175, 191

Microspores, 52

Migration, 119, 138, 169

in birds, 344, 345

Milankovitch, 330

Miller, 18

Missiroli, 102

Mitosis, 46-49, 70, 89

Modification, adaptive, see Homeostasis

Moewus, 256

Mohr, 67, 68

Mollusks, 7, 257, 260

Mongoloids, 159, 354

Monkeys, see Primates

Monoecious, 257; see also Hermaphro-

dites

Monophyletic, 198, 285

Monotremes, 234

Montaigne, 44

Montgomery, 55

Moody, 247

Moore, 144, 178, 179, 184

Morch, 85

Morgan, 36, 53-55, 60, 71, 82, 273, 274

Morgulis, 252

Morphology, 8, 243

Mosquitoes, 103, 229, 230

392

Index

Neumayr, 286

Neurospora, 55, 83, 249

Neutral traits, 152-154

Newman, 241, 278

Niches, ecological, see Ecological niche

Nietzsche, 112

Nomogenesis, 359

Non-disjunction, 62-63

Nordenskiold, 251

Norm of reaction, 74-79, 81, 115, 132,

156, 157, 268; see also Environ-

ment, Heredity, and Phenotype

Notch deficiency, 67

Nucleic acids, 7, 8, 18, 46

Nucleoproteins, 8-16, 18, 19, 21, 46,

73, 80, 371

Nucleus, discovery of, 45

Oakley, 330

Oats, 119, 194, 196, 207, 217

Odontoptera, 104

Oenothera, 82

Oken, 225

Old age, evolutionary, 298

Onion, 48, 100, 154, 196

Ontogeny, 236

Oogenesis, 52, 57

Oparin, 18, 21

Ophryotrocha, 270

Opossum, 293, 295-297, 306, 308, 367,

373

Mutation, useful, 102, 103, 106, 107

viability of, 36, 105-107

see also Bacteria, Biochemical mu-

tants, Diseases, Lethals, and Man

Mutual help, 113

Myths, evolutionary, 111

Naef, 359

Naked genes, 69-70

Narcotics, 13

Natural classification, 247; see also

Phylogeny

Natural selection, 13, 77, 109-133

and adaptation, 19, 91, 102, 106-107,

109-133, 158, 159, 363, 368, 370,

374

and cooperation, 341-344

and culture, 339, 340, 344, 376-378

and ethics, 376-377

and extinction, 367

and isolating mechanisms, 180

and race, 153, 355

and sex genes, 267, 269

and sexual selection, 277-279

and teleology, 230, 231

conservatism of, 130, 131

creativeness of, 115, 130, 131

dynamic, 131

equilibrium with, 143

history of, 111, 112

in man, 122, 153, 359, 374

opportunism of, 113, 364

stabilizing, 131

Nauplius, 238-240

Navashin, 46

Nazi, 112

Neanderthal Man, 330-336, 349

Nearctic fauna, 315, 316

Nectarine, 125

Index

393

Pipilo, 186

Pithecanthropus, see Java Man

Pituitary, 275

Placenta, 277, 293, 306

Placental mammals, 306-309, 367

Plane tree, 188

Plasmagenes, 88, 89

Plasticity adaptive, 345

Platanus, 188

Plato, 134, 225, 227, 253, 370

Pleiotropism, 35-37; see also Gene

Plesianthropus, 327

Pliny, 16, 166, 232, 233

Plunkett, 21

Plutarch, 16

Poa, 64

Pod corn, 210-212

Polar body, 52, 63

Pollen grain, 52

Polydactyly, 32

Polygamy, 259

Polygenes, 40-42, 139

Polymorphism, 136, 138, 142, 143, 163,

219, 220; see also Heterosis

Polyphyletic, 198

Polyploids, 64, 65, 82, 191-221, 260,

284

Pachycephala, 161

Pachytene, 50-52

Painter, 46

Pairing of chromosomes, 52

Palearctic fauna, 315, 316

Paleontology, 284-311; see also Fossil

evidence of evolution

Paludina, 286

Pandorina, 371

Pangenesis, 78-80, 250

Panmictic populations, 115-117, 119,

138

Paracelsus, 16, 166

Parallelism, evolutionary, 306

Paramecium, 89

Parasites, 13-16, 239-240, 367, 371

Parathyroids, 275

Park, 356

Parrot, 340

Parthenogenesis, 270

Pasteur, 1, 17

Patterson, 47, 163, 169, 172

Pavan, 142

Pavlov, 78

Peas, 119, 195

used by Mendel, 25-28, 33, 34, 36

linkage groups in, 55

Peking Man, 328

Penguins, 91

Penicillin, 96, 97

Penrose, 85, 90

Peromyscus, 174

Petersen, 277

Pezard, 275

Phages, see Bacteriophages

Phascolarctos, 307, 367

Phenacodus, 298, 299, 302

Phenotype, 73-77; see also Environ-

ment, Heredity, and Norm of

reaction

Phenyl-thio-carbamide, see PTC

394

Index

Progesterone, 275, 276

Progressive evolution, see Evolution,

progressive

Promptov, 340

Pronephros, 236-238

Prophase, 46-48, 50-52

Prosimii, 324, 325, 335

Protective coloration, 105

Proteins, 7, 8, 11, 14, 19, 80, 125, 246

Przevalsky horse, 197-199

Psyche, 20

Psychoda, 103

PTC, 30, 120, 130, 138, 139; see also

Tasters of PTC

Puccinia, 101, 102

Pure lines, 113-115, 123-125, 136, 189,

258

Pure races, 117, 135, 136, 347

Purkinje, 235

Purposefulness, 109

Quantitative character, 42; see also

Polygenes

Rabbit, 79, 173, 193, 247

Race, definition of,-132^l34=162_^

human. Jg, -135, 136. 138—159-162,

l,

hyl5rtdizalion7~165

in plants, 154-158

major and minor, 159

number of, 160, 347-353

rings of, 182-185

Racial variation, rules of, 158, 159

Racism, 112

Radiation, adaptive, 291-293, 307, 314,

315, 326, 366

damage, 67

Radish, 205-207

Radium, 220

Rana, see Frogs

Ranunculus, 75

Raphanobrassica, 205-207

Raphanus, 205, 206

Rates, evolutionary, 296-298

Rats, 78, 367

Ray, 231

Recessive, 25, 29

Recombination of genes, 30, 53-55, 254;

see also Mendel's laws

Redi, 16

Reeves, 210

Regeneration, 8

Relative sexuality, 255, 256

Remington, 272

Rensch, 158, 168, 362

Replacement, 289

Reproductive isolation, see Isolating

mechanisms

Reproductive success, 119

Reptiles, 234, 291, 295, 313, 363-367

age of, 289-291

Resistance to antibiotics, 96-99

to insecticides, 102-204

to rust fungi, 33

Respiration, 243-246

Retinoblastoma, 139

Reversibility of mutation, 98, 99

Rhesus, 152

Ribonucleic acid, 8, 15

Index

395

Schleiden, 45

Schmalhausen, 131, 133, 371

Schmidt, 356

Schull, 43

Schultz, 59

Schwann, 45

Science, purpose of, 1

Sclaters, 315

Scolytus, 103

Scorpions, 282

Sea squirt, 259

Sea urchin, 45; see also Echinoderms

Secale, see Rye

Sedimentary rocks, 285

Segregation, Mendelian, see Mendel's

laws

Selection, and adaptation, 109-133, 158,

159; see also Adaptation

artificial, 34, 191

coefficient, 119, 120

natural, see Natural selection

sexual, 277-279

speed of, 120, 131

stabilizing, 131

Self-fertilization, 123, 124, 126, 136,

168, 218, 219, 257, 258

Self-pollination, 215, 258

Self-reproduction, 19, 23, 37, 66, 70, 73,

79

Self-sterility, 258, 259

Semilethal, 141, 142

Seminal receptacle, 172

Semon, 79

Senescence, evolutionary, 364-368

Sense organs, 372

Serological reactions, 246, 247

Serotypes, 89

Severino, 242

Sex, and environment, 268-272

as an adaptation, 253

chromosomes, 55-63, 253, 261-266,

270, 274, 281; see also X-chromo-

some, Y-chromosome

determination, see Sex chromosomes

evolution of, 253-283

hormones, see Endocrines

in hymenopterans, 269, 270

in unicellular organisms, 254-256

linkage, 60-67

Sex, reversal, 281

Sexual behavior, 279-281

drive, 275, 276

isolation, see Isolating mechanisms

populations, see Mendelian popula-

tions

reproduction, 23, 35, 38, 70, 91, 131,

132, 254

selection, 277-279

species, 183, 184

Sexuality, relative, 255, 256

Seymouria, 365

Sheep, 81, 173, 193, 248

Shrews, 295; see also Tree shrews

Shull, 215

Sibley, 186

Sibling species, 188, 189

Sick, 340

Sickle cell, 141, 143, 219

396

Index

Spermatid, 50

Spermatocyte, 50, 59

Spermatogenesis, 52, 56

Spermatogonia, 50

Spermatozoa, 10, 11, 34, 45, 52, 222

Sphaerella, 91

Sphenodon, 292, 293, 296, 297, 365,

367

Sympatric, 137, 138, 163, 166-170, 184,

187, 188

Syndrome, 36, 37, 83

Synthetic theory of evolution, 109

Systematics, 181, 187, 189, 190

Tandler, 277

Tarpan, 38, 197, 198

Tarsius, 322, 324, 325, 336, 362

Tasters of PTC, 30, 33, 42, 120, 130,

138, 139

Teeth, 295

Teleology, 230

Telliamed, 166

Telophase, 46, 48, 50

Temperature regulation, 294, 295, 366,

371

Spiders, 282

Spiderworths, 45

Spieth, 172

Spiny anteater, 234, 293

Spontaneous generation, 16, 17, 20

Sports, 81

Stabilizing selection, 131

Stadler, 86

Stakman, 101

Stanley, 14, 15

Staphylococcus, 96, 97

Starling, 274

Stebbins, G. L., 124, 127, 133, 163, 188,

221, 318

Stebbins, R., 186

Steinheim skull, 332

Stenobothrus, 50

Stephens, 204, 205, 221, 248

Sterility, hybrid, see Hybrid sterility

of mutants, 141, 142

Stern, 43, 56, 57, 90

Stevens, 55

Stingless bees, 270

Stone, 47, 163, 169, 172

Strasburger, 46

Streptomycin, 96-99, 106

Struggle for existence, 111, 112

Stud books, 200

Sturtevant, 55, 248

Subspecies, see Race

Sulfonamide, 97

Sunlight, effects of on man, 13, 14, 72,

74, 76, 77, 153, 354

Superfemales, 63, 262, 263, 269

Supermales, 262, 263, 269

Survival of the fittest, 112

Sutton, 53, 55

Swammerdam, 45, 222

Swanscombe skull, 332, 333

Sweet peas, linkage in, 55

Sycamore, 188

shocks, 63

Teosinte, 208-210

Termites, 341, 343

Index

397

Tuatara, see Sphenodon

Tuberous sclerosis, 85

Turtles, 292

Twins, 34, 125, 136, 277, 281

Types, 134-136, 225, 227, 242, 288

Typhoid in poultry, 99-101

Typology, 134, 135

Ultraviolet, biological effects of, 18, 72,

354

Ungulates, 298

Unit characters, 36

Uranium, 286, 287, 359

Urey, 18

Urogenital organs, 236-238, 280

Use and disuse of organs, 77

Useful mutations, see Mutation

Uterus, 275, 293

Vaccination, 100

Vallois, 330

Vanadium, 7

Vandel, 359

Variability, 23, 138, 254; see also Muta-

tion, Race, and Sexual reproduction

Varieties, 138; see also Race

Vavilov, 194, 195, 209

Venereal diseases, 97

Vestigial organs, 240-243

Vetukhov, 178

Victoria, Queen, 85

Violets, 174

Virgil, 16

Virulence, 99-102

Viruses, 13, 73

and genes, 33, 69

bushy stunt, 15

chemistry of, 8, 18

discovery of, 14

evolution of, 70

mutation in, 82

plants, 15

primordial, 19, 69, 371

reproduction of, 16

tobacco mosaic, 15, 33

Vis vitalis, see Vitalism

Vitalism, 13, 19, 20, 79, 273, 357, 359,

369

Vitamins, 13, 35, 76, 83, 153

Vries, de, 25, 82, 83

Waddington, 379

Wagner, 168

Wald, 252

Walker, 100, 108, 154

Wallace, A. R., 278, 311, 315

Wallace, B., 178, 220

Wallace, E., 84, 85, 264

Warm bloodedness, 293; see also Home-

ostasis

Warmke, 265

Washburn, 335

Wasps, 269, 341

Watch analogy, 369

Water crowfoot, 75

Weaver, 75

Wegener, 317

Weidenreich, 333-335, 337, 355

Weier, 75

Weimer, 81, 300, 301

Weinberg, see Hardy-Weinberg law

398

Index

Yak, 174, 193

Yarrow, 154-157

Y-chromosome, in man, 59

sex determination, 261-274

see also Drosophila

Yeasts, 83

Young, 252

Youth, evolutionary, 298, 364-366

Vudkin, 245

Zavadovsky, 275

Zea, see Corn

Zebras, 169, 170, 196, 197, 304, 305

Zebroids, 197

Zeuner, 330

Zimmermann, 228, 313, 318

Zirkle, 22, 79, 90

Zoogeography, 315-317

Zygote, 49, 51, 52, 254